O-lab

olab

Automated tests give you confidence in the piece of code you write. Sometimes we write code that works locally then push to production. However, changes made breaks entirely different parts of the project. Someone else on your team might have written these parts of the project, but your changes caused conflicts. Testing prevents this. There are different types of tests: unit tests, integration tests, functional tests, end-to-end tests, and many more. What is unit testing? Using a car as an analogy, we all know that a car comprises of different components (Unless you live in a tree 🤪). These components comprise wheels, an engine, headlights, and taillights, etc. Unit tests basically ensure each individual component performs as expected. Looking at our car analogy, to ensure our wheel is perfectly round we rotate it by exerting external force, same way we write unit tests to ensure individual components of our source code (software) function as they ought to. In this article, we will be looking at the basic principles of unit tests with a simple react-native component. Button Let’s go ahead to create a component. Button React ; {Pressable, Text, View, StyleSheet} ; Button = { ( <View style={transparent ? styles.transparentBtn : styles.btn}>
        <Text style={transparent ? styles.transparentTitle : styles.title}>
          {isLoading ? 'Loading...' : title}
        </Text>
      </View> );
}; styles = StyleSheet.create({ : { : , : , : , : , : , : , : ,
  }, : { : , : , : , : , : , : ,
  }, : { : , : ,
  }, : { : ,
  },
}); Button; import from 'react' import from 'react-native' const ( ) => {onPress, title, isLoading, transparent} return {
        if (!isLoading) {
          return onPress();
        }
      }}> < = => Pressable onPress {() </ > Pressable const btn width '100%' height 50 backgroundColor '#74B3CE' alignItems 'center' justifyContent 'center' marginVertical 10 borderRadius 4 transparentBtn width '100%' height 50 backgroundColor 'transparent' alignItems 'center' justifyContent 'center' marginVertical 10 title color 'white' fontSize 16 transparentTitle color '#74B3CE' export default Testing We test react-native using several tools. The first tool we will look at is Jest. Jest — Javascript testing framework. There are different testing frameworks, depending on the programming language tested. ‘Jest’ is one of the most preferred testing frameworks for React and React-native applications. There are other Javascript testing frameworks; Mocha, Jasmine, Karma, Puppeteer. React Native ships with Jest; we can skip its installation. ‘Jest’ discovers test files within our project via their file names, which can be at any depth in our project. There are three naming styles by which ‘Jest’ picks up our test files. — This is suitable for finding associated test files for a component. When you have App.js the test file is side-by-side as App.test.js or App.spec.js — This is preferrable for components that require multiple test files and large projects. Any file with a .test.js suffix or a .spec.js suffix Any files witing a test folder Enzyme — Enzyme helps test the output (rendering), interactivity, and manipulation of our React and React native components by simulation, as compiling and running our react-native application for testing would be a slow process. Enzyme’s syntax makes our lives easier while testing components 😁. An alternative to Enzyme is ‘react-test-renderer’, but we will make use of Enzyme as it contains more functionalities. To setup enzyme first run: npm install jest-environment-enzyme jest-enzyme enzyme-adapter-react-16 — save-dev npm install — save-dev react-dom enzyme Edit package.json — Add the following config to “jest” object : { [ ], : , : { : }  
    } "jest" "setupFilesAfterEnv" "jest-enzyme" "testEnvironment" "enzyme" "testEnvironmentOptions" "enzymeAdapter" "react16" So lets get to it! Testing our component Button React ; Button ; {View, Text} ; {shallow} ; toJson ;

describe( , () => {
  it( , () => { component = shallow( import from 'react' import from '../index' import from 'react-native' import from 'enzyme' import from 'enzyme-to-json' 'Button' 'should render without issues' const );

    expect(component.length).toBe(1);
    expect(toJson(component)).toMatchSnapshot();
  });

  //... < /> Button , , - Are global variables in ‘Jest’; as a result, we don’t have to require or import anything to use them. describe it expect - This method creates a block or scope where we write several related tests. 'describe' takes in a name (should be descriptive) and then a callback where we write our actual tests describe(name, cb) - ‘it’ is an alias of test which has three parameters — a descriptive name of your choice, a callback function containing our test assertions, and an optional timeout that specifies how long to wait before the test aborts. The default timeout is 5 seconds. it(name, cb, timeout) - Expect is used when we want to test a value, we mostly use 'expect' alongside other functions called "matchers", to view all matchers you can check the link expect(value) Expect · Jest Testing “Button”, we call the method passing in a description and a callback indicating a shared relationship of every test assertions in the describe block. Then our first test assertion — makes sure the renders. describe it(‘should render without issues’, ….) Button Using from Enzyme, we run the component in isolation without rendering its children (In this instance, Button has no child component), calls the constructor of the component and the component’s method. shallow Button shallow render The number of components rendered (component.length) is expected to equal one using and the matcher. compares values or object instances, making it more suitable to the operator. expect toBe() toBe() === —We use to make sure we don’t cause unexpected changes to UI components. When we make deliberate changes to components, the test fails; as a result, we update our previous using Snapshots snapshots snapshots snapshots npm test — -u We use the to compare previous and current snapshots. When there isn’t an old snapshot, ‘Jest’ creates a new one. .toMatchSnapshot() To compare snapshots we use the method; it returns a compatible format of our component for ‘Jest’ snapshots. toJson( component ) The next thing we want to test is the event prop, but before doing, that let’s look at Mock functions in Jest. onPress — In Jest, mock functions allow us to test links between code. Jest mock functions let us erase the actual implementation, more like a placeholder. For example, testing a component that makes actual API calls, our test will be slow; hence we mock the API endpoint. Mock functions There are two ways to mock functions in ‘Jest’: Creating in test scripts.Writing to override dependencies. mock functions manual mocks For now, we will mock up a simple function below. We track of mock functions calls and return values; using the property. This is done using the , , properties. .mock .calls .return .instances it( , () => { result = ; mockFn = jest.fn( result); component = shallow( //... 'should call onPress event' const 'I was Pressed' const => () const );

    expect(mockFn).not.toHaveBeenCalled();

    component.props().onPress();

    expect(mockFn.mock.calls.length).toBe(1);
    expect(component.props().onPress()).toBe(result);
  });

  //... < = /> Button onPress {mockFn} We create a mock function which returns ‘ , then render our component using Afterwards simulate a press by calling . The method returns an object containing all props of a component. Considering the normal behavior of a button, pressing the component only once should call our mock function — which returns ‘ To test for this, is called twice Ensuring is called once, and its return value is correct. I was Pressed’ shallow(). props().onPress(). .props Button I was Pressed’. expect() . mockFn The next thing we want to make sure of is not calling when is onPress mockFn isLoading true . it( , () => { result = ; mockFn = jest.fn( result); component = shallow( //... 'should not call onPress event when loading is true' const 'I was pressed' const => () const );

    component.props().onPress();
    expect(mockFn.mock.calls.length).toBe(0);
  });

//... < = = /> Button isLoading {true} onPress {mockFn} When prop is , we simulate a press on Button, which shouldn’t trigger a call to mockFn. Therefore the number of calls to mockFn remains 0. isLoading true Next, we test if renders the correct title. Button it( , () => { title = ; component = shallow( //... 'button should render title passed to it' const 'Primary Button' const );

    expect(component.find(Text).first().props().children).toBe(title);
  });

//... < = /> Button title {title} We use the method to obtain the Text component in , the child of should be the same as the prop. .find Button Text title Next, we test if the title of renders when Button ‘Loading…’ isLoading=== true it( , () => { component = shallow( //... 'should be Loading... text when isLoading === true' const );
    component.setProps({isLoading: true});
    component.update();
    expect(component.find(Text).first().props().children).toBe('Loading...');
  });

//... < = /> Button isLoading {false} We use to update the props passed to and force a re-render, then we check if the title rendered is setProps Button “Loading…” Next, we test if the styles of change when we pass the prop to Button transparent Button. it( , () => { transparentContainer = { : , : , : , : , : , : ,
    }; transparentTitle = { : ,
    }; component = shallow( //... 'should have transparent styles when transparent is true' const width '100%' height 50 backgroundColor 'transparent' alignItems 'center' justifyContent 'center' marginVertical 10 const color '#74B3CE' const );
    expect(component.children(View).props('style').style).toMatchObject(
      transparentContainer,
    );

    expect(
      component.children(View).children(Text).props('style').style,
    ).toMatchObject(transparentTitle);
  });
}); < /> Button transparent We use the to check if our style objects are correct. .toMatchObject Finally, we test by running npm run test Our test cases pass, yeee! 😝 Conclusion Testing is like our best friend, saving us from a lot of troubles. Testing our code prevents us from pushing breaking changes to production, forces us to write quality code, serves as a form of documentation, find bugs early, and a lot of other advantages. Resources For further readings and API references, check out the links below: Jest · 🃏 Delightful JavaScript Testing Testing · React Native Testing React Native Apps · Jest Also published at: https://medium.com/react-native-nigeria/getting-started-with-unit-testing-in-react-native-1fb2d0add259 Lead Photo by on Alvaro Reyes Unsplash

Beginners Guide to Get Started with Unit Testing in React Native

Idea / inspiration I’m creating a project that can spot trees, using satellite data. I got this idea from the company . Which uses satellite imagery and AI to track forests and vegetation around the world. They are many uses for this technology. You can track deforestation in an area. By tracking the amount of missing trees over a certain timescale. A company called uses satellite imagery and AI. To prevent fires and power outages by spotting trees growing too close to power lines. 20tree.ai Spacept Using satellite imagery saves time and money. As inspecting forestry using a helicopter or drone can cost a lot of money. And will cost a lot of time, to set up the helicopter and drone to fly over the forestry. Dataset When I was first trying to get the dataset together for the machine learning model.  I settled on using . I got this idea from this . Which used satellite imagery to spot landslides. In the blog post he highlighted the issue of getting satellite data. Where if you wanted to get images from the . You will have to do them manually. And you cannot add coordinates of an area you want to capture. google earth engine medium blog post ESA Copernicus SciHub website So, you get continent sized images. Which I can agree from my personal use of the service as well. When I tried to get an image that covered a small area in my city. Instead I got large images covering all of France and Britain. With Google earth engine you can filter by date, area, and cloud cover. Which is why Google earth engine was recommended.  I used the code from the blog to create a dataset. I added customisations to the code for my dataset. The blog post described the process below: How our final (semi) automated pipeline ended up working was as follows: 1. Enter coordinates intocropping_coordinates.py to calculate the coordinates of a 10 km square. 2.  Copy the output and replace the variables at the top of the JavaScript code in the Google Earth Engine console (the JavaScript code we used). Don’t forget to change the date range to one that you’d like. 3.  Download the images with less than 100 for the mean cloud density given by clicking the link printed in the console output. When I got the images, they looked like this: Image by author In the beginning I did not know why this was the case. But later learned that satellite images were separated into different bands. Bands are certain wavelengths that the satellite captures. Doing this allows the user to view the image with features that are not visible with human eye or can be highlighted with a chosen wavelength. For example, a common use of satellite imagery is tracking farmland. By using the infrared spectrum, it is more easier spot vegetation. Figure 1 This false-colour image of Florida combines shortwave infrared, near infrared, and green light. (NASA by Matt Radcliff with data from the USGS ) image Landsat 5 Earth Explorer. Also found here: https://earthobservatory.nasa.gov/features/FalseColor gives a nice list of wavelengths and their uses: Wikipedia The wavelengths are approximate; exact values depend on the particular satellite's instruments: Blue, 450–515..520 nm, is used for atmosphere and deep water imaging, and can reach depths up to 150 feet (50 m) in clear water. Green, 515..520–590..600 nm, is used for imaging vegetation and deep water structures, up to 90 feet (30 m) in clear water. Red,600..630–680..690 nm, is used for imaging man-made objects, in water up to 30 feet (9 m) deep, soil, and vegetation. Near infrared (NIR), 750–900 nm, is used primarily for imaging vegetation. Mid-infrared (MIR), 1550–1750 nm, is used for imaging vegetation, soil moisture content, and some forest fires . Far-infrared (FIR), 2080–2350 nm, is used for imaging soil, moisture, geological features, silicates, clays, and fires. Thermal infrared , 10400-12500 nm, uses emitted instead of reflected radiation to image geological structures, thermal differences in water currents, fires, and for night studies. Radar and related technologies are useful for mapping terrain and for detecting various objects. Also, : combinations True-color uses only red, green, and blue channels, mapped to their respective colors. As a plain color photograph, it is good for analyzing man-made objects, and is easy to understand for beginner analysts. Green-red-infrared, where the blue channel is replaced with near infrared, is used for vegetation, which is highly reflective in near IR; it then shows as blue. This combination is often used to detect vegetation and camouflage. Blue-NIR-MIR, where the blue channel uses visible blue, green uses NIR (so vegetation stays green), and MIR is shown as red. Such images allow the water depth, vegetation coverage, soil moisture content, and the presence of fires to be seen, all in a single image. I wanted a simple RGB image. Which is simple visible light. As its best to start simple. To develop that, we need an image that combines all 3 bands into one. So, using some of the google earth engine example code and my personal changes. I was able to get some images. The red square covering the map is the area that google earth engine will capture for the image. Image by author The image below: Image by author As we can the image quality is horrible to be frank. I started to notice problems when setting up the red square. You see in the image of the general area of outside the red square. You can see that image lacks clarity. Which means if you zoom in even further you cannot tell what’s in the image because the quality is pretty bad. The reason for the problem was the resolution of images. The satellites sentinel-2 and Landsat only have a resolution 30 meters. This explains spatial resolution perfectly: website Spatial resolution refers to the size of one pixel on the ground. A pixel is that smallest ‘dot’ that makes up an optical satellite image and basically determines how detailed a picture is. Landsat data, for example, has a 30m resolution, meaning each pixel stands for a 30m x 30m area on the ground. It’s considered a medium-resolution image, which can cover an entire city area alone, but the level of detail isn’t fine enough to distinguish individual objects like houses or cars. Custom Dataset All open source data goes down to 30 meters. For finer data one must pay. So I had to find a solution that allows me get higher quality images. Without forking small fortune. After a lot thinking (and worrying). I found the solution hidden in plain sight. When using navigation systems on your phone. Your phone can give you a satellite image of your area. You can see which buildings are which. And differentiate between objects. So I decided to use (normal) google earth and capture images straight from the website. To make sure the screen captures where all the same size used a program called PicPick. Here’s one image I captured: Image by author Labelling data Now we have a small sample size. We need a solution for the machine learning model to know it ‘s looking at a tree. So used a program called Which I used to manually label the trees in the images. LabelImg from GitHub. Image by author I did not label every single tree in every image. As that will take forever. But wanted to have some labelling so the model can pick one or two trees in the test image. Same reasoning why the dataset is very small. The reasoning is so if I need to make changes to the dataset. It won’t take a long time. Also, just to create a model as simple prototype before scaling up the work. Model This project is an object detection problem. As I’m dealing with image data. I will be using some type of Convolutional Neural Network. One example I was thinking of using for the model is . But quickly learned that as the model was already trained the model won’t be able to read satellite image for trees. The blog post mentioned you do transfer learning or fine-tuning. But said t that it was outside the scope of the article. So, I had to find a new model to use. here R-CNN After a while of googling around. I found an interesting model that I can use called R-CNN. More precisely faster R CNN. R CNN which stands for Region-Based Convolutional Neural Network. Where the model selectively goes through many 2000 proposal regions of the image. After the proposal regions are selected, they then shaped for the input of the Convolutional Neural Network. After Support vector machines classify the regions. Then a bounding box regression is used to work out where the box goes. Figure 2 https://www.analyticsvidhya.com/blog/2018/10/a-step-by-step-introduction-to-the-basic-object-detection-algorithms-part-1/ But they are a few problems relating to using a traditional R CNN. Mainly that processing an image takes a lot of time. As the model is extracting 2000 regions to check. Prediction with the model takes around 40 seconds. Fast R-CNN A better version of R CNN was created by the same person. Called Fast R CNN. Main difference is instead of extracting thousands of regions from the image. Fast R-CNN inputs the whole image. Then the CNN produces a convolutional feature map from that the chosen region proposals. The region proposals are then shaped for the pooling layer. After that it gets reshaped again to be 4entered into a full connected layer. For each region of interest feature vector, it is moved to a SoftMax layer to predict the class. The use of bounding box regression to find the boxes. One problem is that as Fast R-CNN uses selective search is slowing down testing. But Fast R-CNN is way faster than the traditional R-CNN. As Fast R-CNN takes around 2 seconds to predict. Faster R-CNN Now we have faster R-CNN. The examples above uses selective search to go through proposal regions. Faster R-CNN uses a separate network of predict the region proposals. Using the convolutional feature map. Then the proposals are then reshaped for a pooling layer. For classification of the region. Also bounding box regressor used to place the images. Figure 3 https://www.analyticsvidhya.com/blog/2018/10/a-step-by-step-introduction-to-the-basic-object-detection-algorithms-part-1/ Coding up the of model I followed tutorial from . The tutorial used a dataset of blood cells. I replaced the dataset with my custom dataset of google earth photos. To get the data ready for the model I had to convert the bounding boxes from XML to CSV. here To do that I used pandas and python’s ElementTree module. And borrowed code from and the export file in the tutorial. machine learning mastery linked list_dataframe = []
folder = file os.listdir(folder):
    path = file = path + file
    print(file) filename = re.sub( , , file)
    filename = filename.replace( , )
    filename = filename.replace( , )
    tree_type = tree = ElementTree.parse(file) root = tree.getroot() boxes = list() box root.findall( ):
        xmin = int(box.find( ).text)
        ymin = int(box.find( ).text)
        xmax = int(box.find( ).text)
        ymax = int(box.find( ).text) row_list = [filename, tree_type, xmin, xmax, ymin, ymax]
        print(row_list) list_dataframe.append(row_list)


print(list_dataframe)

df = pd.DataFrame(list_dataframe, columns=[ , , , , , ])

print(df.head())

df.to_csv( , index= ) 'Custom_dataset\\Labels' for in 'Custom_dataset\Labels\\' # filename = os.path.splitext(file)[0]+'.png' 'Custom_dataset\\\Labels' "" "\\" "" ".xml" ".png" 'Tree' # get the root of the document # extract each bounding box for in './/bndbox' 'xmin' 'ymin' 'xmax' 'ymax' # coors = [xmin, xmax, ymin, ymax] # boxes.append(coors) # df = df.reset_index(drop=True) 'filename' 'tree_type' 'xmin' 'xmax' 'ymin' 'ymax' 'testing.csv' False df = pd.DataFrame( , columns=[ , , , , , ])
list_dataframe = []
folder = None 'filename' 'tree_type' 'xmin' 'ymax' 'ymix' 'ymax' 'Custom_dataset\\Labels' This section sets up the variables to be used for the loop. An empty dataframe is set up with the name of the filenames, class and the bounding boxes. The path of folder containing the images are saved in the folder variable. file os.listdir(folder):
  path = file = path + file
  print(file) for in 'Custom_dataset\Labels\\' The start of the for loop starts by listing the files in the folder. Then the file path then printed by combining the folder path and name of the file. This is done as a normal for loop without os.listdir did not work on my laptop. filename = re.sub( , , file)
filename = filename.replace( , )
filename = filename.replace( , ) 'Custom_dataset\\\Labels' "" "\\" "" ".xml" ".png" When printing the filename it had too many backslashes. So to get rid of it str.replace functions were used. Also, as the folder it using the label names not the images. The filenames ended in XML. So, replaced to PNG filetype. tree_type = tree = ElementTree.parse(file) root = tree.getroot() boxes = list() 'Tree' # get the root of the document # extract eachbounding box The section parses through the xml file. Crates the name of the class. In tree_type. tree = ElementTree.parse(file) parses the file. Then root = tree.getroot() gets the root of the file. And saves bonding box in a list. box root.findall( ):
     xmin = int(box.find( ).text)
     ymin = int(box.find( ).text)
     xmax = int(box.find( ).text)
    ymax = int(box.find( ).text)

   row_list = [filename, tree_type, xmin, xmax, ymin, ymax]
   print(row_list)
   List_dataframe.append(row_list) for in './/bndbox' 'xmin' 'ymin' 'xmax' 'ymax' The for loop goes through each of the boxes and extracts the coordinates. Of the four corners of the box. Then a list is created which saves the filename, class, and coordinates. Then appended to the dataframe from earlier. print(list_dataframe)
	
df = pd.DataFrame(list_dataframe, columns=[ , , , , , ])
	
print(df.head())
	
df.to_csv( , index= ) 'filename' 'tree_type' 'xmin' 'xmax' 'ymin' 'ymax' 'testing.csv' False After that list_dataframe containing all the data of the files. Is then added into another dataframe with the column names. And df.to_csv turns the dataframe into a CSV file. Implementing the model Now the tutorial asks for use to convert the CSV file into txt file. As the model uses a different format. filepath, x1,y1,x2,y2,class_name Where: filepath is the path of the training image x1 is the xmin coordinate for bounding box y1 is the ymin coordinate for bounding box x2 is the xmax coordinate for bounding box y2 is the ymax coordinate for bounding box class_name is the name of the class in that bounding box pandas pd

data = pd.DataFrame()
data[ ] = train[ ] i range(data.shape[ ]):
    data[ ][i] = + data[ ][i] i range(data.shape[ ]):
    data[ ][i] = data[ ][i] + + str(train[ ][i]) + + str(
        train[ ][i]) + + str(train[ ][i]) + + str(train[ ][i]) + + train[ ][i]

data.to_csv( , header= , index= , sep= ) import as 'format' 'image_names' # as the images are in train_images folder, add train_images before the image name for in 0 'format' 'train_images/' 'format' # add xmin, ymin, xmax, ymax and class as per the format required for in 0 'format' 'format' ',' 'xmin' ',' 'ymin' ',' 'xmax' ',' 'ymax' ',' 'cell_type' 'annotate.txt' None None ' ' Now the tutorial asks us clone this When I ran the model I got a few errors. So, I went inside the repository and edited the code. Most errors were AttributeError: module 'keras.backend' has no attribute 'image_dim_ordering'. I found this helpful solving the problem. This is because some of the code used is deprecated. And you need to use the new keras API which has different names for the functions. GitHub repository. GitHub issue answer So, after a few hours of googling other issues the machine learning model started to run. I used a google cloud deep learning VM to train my model. Image by author Due to the amount of time. A decided to reduce the amount epochs to 5. As each epoch took 4 hours. And the default number was 1000. Which will take couple of days to train and quite a bit of money. Now I needed to test model. I did this on my laptop. As Google’s VM did not show images To do this I ran the statement: python test_frcnn.py -p test_images Image by author The output shows it found some trees: Images by author The model was able to find some areas with trees. Mainly thick forestry. But others, not at all: Images by author It looks like the model can’t point individual trees. But it can point to big areas forestry. Improvements Like I mentioned above the model can’t spot individual trees. But areas of forestry. But even then with the areas of forestry. It can only spot some sections. A few small boxes cover a large area of forestry. The dataset was kept brief. So, I can get a workable prototype up and running. Labelling can be improved by boxing every single tree in an image. And having custom segmentation for areas forestry. Instead of boxes to cover an area of forestry. The dataset is very small (20 images) and trained for only 5 epochs. This can be improved by simply adding more images to the dataset. And spending more time training the model. A serious question to ask is, how I would label a large dataset. Labelling 20 images was already long enough. By manually labelling the trees. Took around hour or so. Can’t imagine how to label thousand images by hand. Also, with custom segmentation for large areas forestry I’m not sure how to make custom shapes. To cover the shape of the forestry. As the program I was using only allows for boxes. Not custom shapes like polygons. Published from my website

How To Identify Trees with Deep Learning

If the complex and costly hardware required for traditional mining has been a deterrent for you, or if you’re weary of the relentless grind to boost transaction volumes just to earn fees, now there’s a refreshing alternative. Discover the brand-new mobile mining experience that ‘ ’ provides, where your smartphone becomes your portal to a rewarding ‘verify to earn’ journey. Massive Introduction We’re thrilled to introduce , our innovative mobile mining app, poised for launch alongside the upcoming at beginning of April. As the first project in the blockchain sector to leverage ‘ZKVM+DePIN+AI’ technology, Ola is at the forefront, pioneering a new, fair, open and user-centric approach to blockchain interaction with this initiative, Massive. Massive Ola Testnet Alpha Massive simplifies the blockchain experience by removing the need for expensive hardware or technical expertise. Our vision for Massive is direct and clear: to democratize web3 access by providing a platform where anyone with a smartphone can participate and earn rewards through simple verification tasks. Our commitment with Massive is to lower the entry barriers to blockchain technology, making it accessible and equitable for all users. By introducing the ‘verify to earn’ concept, we’re not only changing how people mine; we’re also transforming their role within the blockchain ecosystem from passive observers to active participants. Massive, as a crucial part of Ola’s development strategy, . Ola is set to open applications for invitation codes on March 18th, releasing the first set of 1,000 codes for node registration. This initial offering is limited to 1,000 nodes, with registrations being accepted on a first-come, first-served basis. Additionally, the first 1,000 nodes will have the privilege of generating invitation codes, allowing them to invite newcomers with these codes to earn rewards. plans to allocate about 10% of Ola’s token to verifiers & miners, with an initial 1% being used for Ola Testnet Alpha Join us as we embark on this exciting journey to make blockchain technology intuitive, inclusive, and rewarding for everyone, everywhere. Background The industry has tirelessly explored various strategies to engage traditional users (those outside the web3 ecosystem), including gaming, infrastructure, and incentivizing through airdrops. Notable efforts like StepN’s ‘move to earn’ and Starkware’s airdrops for non-Web3 developers momentarily captured high interest, hinting at potential breakthroughs. However, their inability to sustain long-term momentum and impact highlighted the critical need for innovative, enduring approaches that truly resonate with and retain users in the blockchain realm. “We recognized the need for a standout feature that resonates deeply with users; thus, Massive was conceived as our bold solution to this industry challenge,” - Simon Chen, co-founder and CEO of Ola Ola’s mission with Massive is to catalyze the widespread adoption of web3, advocating for a decentralized and fair digital ecosystem. By introducing the ‘verify to earn’ concept, where anyone with a smartphone can participate in mining, Massive aims to significantly lower these barriers. Additionally, users can engage with multiple projects over time on their smartphones, enhancing the longevity and diversity of their participation. This approach will foster a more inclusive environment, encouraging a broader demographic to engage with, understand, and contribute to the web3 space, while receiving long-term rewards. The introduction of Massive is a strategic move toward making blockchain more accessible and engaging for the general public, thus infusing the industry with new participants and fresh perspectives. Through this initiative, Ola envisions transforming mere observers into active contributors, nurturing a dynamic and sustainable blockchain community. Massive: Fair Opportunities for All In the realm of traditional mining, such as the Proof of Work (POW) mining mechanism, the requirement for costly hardware presents a significant barrier to ordinary users. When developing Massive, Ola focused on reducing entry barriers, enhancing user experience, and increasing user rewards, all while guaranteeing utmost fairness for EVERY user. Massive is a decentralized authentication network composed of various mobile elements. In terms of user-friendliness, Massive boasts several key features: User Access Threshold: : Users can easily access Massive by simply downloading the necessary app and wallet on their smartphones. Low Entry Barrier : Massive supports a broad range of mobile models, compatible with both Android and iOS (coming soon) systems. Minimal Device Requirements User Rewards (10% of Ola’s tokens): : Users receive rewards by completing verification tasks, providing a straightforward way to earn. Verify to Earn : In addition to earning verification rewards, users can also gain additional returns by inviting newcomers with a referral code and through staking activities. Multiple Reward Streams User Experience: : Users can join or leave the network at their convenience, offering the flexibility to suit individual needs. Flexible Access and Exit : Without the need for complex operations or maintaining high transaction volumes, Massive simplifies user participation. Ease of Use : Compared to distributing funds across multiple addresses, using a smartphone to participate presents a lower barrier for the average user. User-Friendly Enhanced Network Security and Intelligence: : Massive harnesses AI+zk technology to elevate network security and intelligence, utilizing AI for nuanced analysis of node service statuses. This results in refined task distribution and heightened network efficiency. AI + ZK Integration Platform Fairness and Sustainability: : Massive ensures fairness for all users, where everyone’s time and contributions are equally valued. Fair : Users can engage with multiple projects over time on their smartphones, enhancing the longevity and diversity of their participation. Sustainable Massive Changes the Game: Say Hi to Your New Web Roles At Massive, your role is fundamentally transformed. Gone are the days of having to “slave away” by artificially boosting transaction volumes and amounts to earn incentives. Now, all you need is a standard smartphone to connect. By simply staying active and verifying, you can consistently earn rewards. But your role evolves further; once you connect as an Ola node, you become an essential part of maintaining the network’s security. You might have felt the imbalance before, as web3 whales could distribute their funds across multiple addresses to amplify their earnings, a feat not so easily replicated by purchasing multiple smartphones. Hence, Massive levels the playing field, offering a balanced opportunity for both whales and retailers. On the Massive platform, everyone participates and contributes in the same manner, ensuring fairness. This approach incentivizes a broader spectrum of users to engage with the web3 ecosystem, lowering entry barriers and bolstering decentralization and equity within the system. With Massive, every participant has the chance to earn rewards fairly, fostering wider participation and a healthier network. Distinct from other staking or mining models where more staking could mean more chances to validate, Massive champions decentralization and equity. Every node stakes an equal value, and by merely staying online and completing verification tasks, rewards can be earned. Time is valued equally across all participants, ensuring that mining on Massive is inherently fair. This paradigm encourages broader and more equitable user engagement, reinforcing the network’s integrity and community inclusiveness. Looking Ahead: Shaping the Future of Blockchain and Web3 Mobile Experiences with ZK Technology As we gaze into the future, our expectations for the Massive are set even higher: Firstly, as a purely decentralized ZK verification network, Massive is poised to accelerate the modularization of blockchain design. Secondly, with an increasing influx of new users joining the blockchain, we anticipate the emergence of genuine Web3 mobile technology. This evolution promises not only to propel the advancement of blockchain technology and invigorate the ecosystem but also to offer users a more equitable, open, and convenient digital experience. Collectively, these efforts aim to facilitate the widespread adoption and development of blockchain technology, steering the industry toward a more prosperous future. Join the Venture! Ola is tackling blockchain’s biggest hurdles: fairness and accessibility, with its gamechanging ‘Massive’ app, leveraging “ZKVM+DePin+AI” technology. Massive simplify blockchain interactions with a ‘verify to earn’ model, opening the digital door for anyone with a smartphone. 🗓️ March 18 marks the launch of Ola’s first 1,000 node registration codes, with mining, new features, and greater rewards soon to follow. Early nodes gain the advantage of creating invite codes for collaborative reward earning. Join us on this venture and be among the first to benefit! About us is an open-source Hybrid ZK Rollup that delivers programmable scalability and privacy to blockchains, empowering individuals through data ownership. Incubated by cryptographers and engineers of Sin7y Labs in 2022, Ola offers a layer2 ZK Stack, featuring Ola-lang, OlaVM, OlaUI, OlaOS, and a series of developer-friendly tools, facilitating the development of any type of contracts. In 2023, Ola launched the Ambassador Program and OVP Points Rewards, now with 20+ global ambassadors. For 2024, Ola plans to roll out public testnet, mainnet, mining mechanism (PoW, ZKP, & Verifier), and a range of incentive programs steadily. If you are interested in joining or partnering with us to contribute to a fair, decentralized, and permissionless world, please contact us at . Ola contact@olavm.org | | | | | | | | | | Website Twitter Discord Telegram Whitepaper GitHub LinkedIn YouTube HackMD Medium HackerNoon

The First Ola Mobile Mining App: A New ‘Verify to Earn’ Paradigm with a 10% Token Airdrop Plan

প্রথম ওলা মোবাইল মাইনিং অ্যাপ: একটি 10% টোকেন এয়ারড্রপ প্ল্যান সহ একটি নতুন 'ভেরিফাই টু আর্ন' দৃষ্টান্ত

Die erste mobile Mining-App von Ola: Ein neues „Verify to Earn“-Paradigma mit einem 10 %-Token-Airdrop-Plan

La primera aplicación de minería móvil de Ola: un nuevo paradigma de "verificar para ganar" con un plan de lanzamiento aéreo de tokens del 10%

La première application Ola Mobile Mining : un nouveau paradigme « Vérifier pour gagner » avec un plan Airdrop de 10 % de jetons

पहला ओला मोबाइल माइनिंग ऐप: 10% टोकन एयरड्रॉप प्लान के साथ एक नया 'वेरिफाई टू अर्न' प्रतिमान

最初の Ola モバイル マイニング アプリ: 10% トークン エアドロップ プランを備えた新しい「検証して獲得」パラダイム

최초의 Ola 모바일 채굴 앱: 10% 토큰 에어드롭 계획을 통한 새로운 '수익 확인' 패러다임

O primeiro aplicativo de mineração móvel Ola: um novo paradigma de ‘verificar para ganhar’ com um plano de lançamento aéreo de token de 10%

Первое мобильное приложение для майнинга Ola: новая парадигма «подтверди, чтобы заработать» с планом Airdrop 10% токенов

İlk Ola Mobil Madencilik Uygulaması: %10 Token Airdrop Planıyla Yeni Bir "Kazanmak İçin Doğrula" Paradigması

Ứng dụng khai thác di động Ola đầu tiên: Mô hình 'Xác minh để kiếm tiền' mới với kế hoạch thả 10% mã thông báo

第一个 Ola 移动挖矿应用程序：新的“验证赚取”范式和 10% 代币空投计划

Welcoming the New Year, we're excited to announce that the Ola Incentivized Pre-Alpha Testnet is now live, inviting both developers and users to foster engagement and innovation. , as an open-source Hybrid ZK Rollup, enhances blockchain ecosystems with programmable scalability and enhanced data privacy. This inaugural testnet release includes features like Ola-lang 0.1.0, OlaOS 0.1.0, AA Deployment, ECDSA & Poseidon Hash, among others, facilitating the development of any type of contract. Ola Ola will conduct rounds of incentivized public testnets leading up to the Mainnet launch. A total of 10 million future Ola tokens, limited-edition NFTs, Olavm Points (OVPs), and mystery rewards will be allocated as incentives. can create innovative zk dApps in just 4 steps using the comprehensive Ola toolkits, which include Ola-lang 0.1.0, OlaVM, OlaUI, and OlaOS 0.1.0, with further advancements on the way. Developers are empowered to actively participate in Ola community governance. All OVP holders will have the opportunity to propose and vote on Ola's future direction, shaping the evolution. Users Join us on this transformative journey, where together we'll unlock Ola's immense potential and share in the rewards of our collective success. Ola Incentivized Pre-Alpha Testnet Overview Easy Building with Ola: A Developer's Guide 🛠️  Tools and Functions The Ola Incentivized Pre-Alpha Testnet provides developers with a minimum viable version of the Ola network. Developers can write smart contracts based on their business needs, deploy them on the Ola network, and use command-line tools and SDKs to send transactions and query status. More features and tools will be released in the near future. The features supported in this release include: : Ola-lang is a Turing-complete language that can be used to write provable programs. The syntax of Ola-lang is designed to be similar to Solidity, making it convenient for smart contract developers to get started. You can find the Ola-lang developer . Ola-lang 0.1.0 documentation here : OlaOS is the implementation of the Ola node. Users can submit transactions, query data, and check transaction status through the JSON-RPC interface to OlaOS. OlaOS 0.1.0 : Ola natively supports Account Abstraction(AA). Users can deploy custom AA contracts and handle signature and transaction execution logic through AA. AA Support : Ola provides support for ECDSA and supports the zk-friendly hash function, Poseidon hash. ECDSA and Poseidon Hash 🖥️  Deploy Your ZK dApp in 4 Steps The Ola DevTestnet offers a local development environment designed for quick and efficient building and testing of Ola smart contracts. The setup process – encompassing environment configuration, account creation, smart contract deployment, and interaction – is streamlined to be completed in under 20 minutes. For a detailed step-by-step guide, visit our tutorial at . Ola Developer Quick Start Setting Up the Environment To begin interacting with the Ola network and compiling Ola-lang code, the command-line tool Olatte needs to be installed. Properly configuring environment variables for Olatte simplifies command execution and improves readability. Installation instructions are tailored for different operating systems: For macOS: via Homebrew For Debian/Ubuntu: via APT For Red Hat/Fedora: via RPM Creating an Account Once Olatte is installed, you can proceed to create a signer. This involves creating a keystore file and registering a public key, essential steps for secure interactions on the network. Deploying a Smart Contract Writing and compiling Ola smart contracts is straightforward with Ola-lang. Use the following command to compile: Lua
luaCopy code
olatte compile path/to/MyContract.ola path/of/output/dir This will generate a .bin file for deployment and an .abi file for interface definitions. Deploy your contract with: Bash
bashCopy code
olatte deploy --network test path/to/MyContract_bin.json Interacting with a Smart Contract Olatte facilitates two primary methods for smart contract interactions: call: Executes read-only functions. invoke: Performs write operations that alter the contract's state. With these simple steps, you'll have the Ola testnet operational in just a few minutes, ready for development and testing! 🤔 Advantages and Opportunities Ola stands out in ZK dApp development, offering scalability, enhanced privacy (coming soon) through zero-knowledge proofs, and an intuitive toolkit for both new and seasoned developers. In addition, Ola supports innovation with incentives for promising projects, providing financial and resource backing. As the testnet progresses enthusiastically, we will collaborate with top global developer communities and conduct online workshops. Our core team members will lead developers to explore Ola, deploying a variety of interesting dapps on our testnet using Ola's toolkits. Developers can explore these opportunities in and apply for incentives at: Ola Builder Program https://forms.gle/AFM4h3Jc5Y23Xfr28 Fair Participation Prioritized: Toward a Decentralized Ola Community Hub 👥  Advancing Community Empowerment Ola is committed to enhancing data privacy and ownership, with a strong focus on community-driven growth. The launch of the and in 2023, featuring over 20 global ambassadors, fosters active participation in shaping Ola’s future. This strategy aims for balanced development, gradually transitioning towards a more decentralized ecosystem. OVP Program Ambassador Program Community members, including managers and content creators, play a growing role in decision-making processes, while network maintainers ensure stability and security. Strategic partnerships and adherence to regulatory standards are also crucial to Ola’s growth strategy. This collaborative and decentralized approach is expected to cement Ola’s position in the blockchain infrastructure, promoting long-term success through community engagement and technological reliability. Visit Ola community hub for more information: https://ola-2.gitbook.io/ola-community-hub/ 🗳 ️Coming Soon: Vote and Propose to Engage in Ola’s Future We’re issuing limited OVPs and empowering all #OVP holders to participate in governance and decision-making. At Ola, we value our community’s support and engagement, offering limited-quantity OVPs as appreciation. By empowering each OVP holder with the opportunity to actively shape our future, we’re fostering a strong community. In the coming weeks, we’ll invite all OVP holders to voice their opinions in our governance and decision-making processes. Stay tuned to make a significant impact on the Ola ecosystem. 🤑 Join Ola X Aspecta Galxe event and share a growing OVP prize pool With the launch of the Ola Pre-Alpha testnet, we are excited to launch Ola BuilderState on Aspecta and celebrate alongside Aspecta on Galxe. All developers and users are welcome to participate, share in a 500k OVP pool and obtain an Aspecta identity. Ola and Aspecta have established a long-term partnership to elevate the global developer community and extend ZK expertise. Looking ahead, we’re facilitating knowledge sharing and hosting a series of reward-given developer-centric events and campaigns to foster a dynamic, innovative, and cooperative ecosystem for everyone involved. Ola Future Ahead: Upcoming Features & Roadmap Our post-Devtestnet roadmap is geared towards testnet iteration, implementing crucial upgrades, and introducing a range of new features and functionalities to enhance Ola, ultimately culminating in a fully decentralized mainnet that offers optional privacy features. Upcoming features in the internal testing phase, expected to be part of the next stage, include: : Introduce mining mechanisms that secure the network and incentivize participants. PoW, ZKP, and Verifier Mining Mechanisms : To maintain network integrity and prevent DDoS attacks, each transaction will require gas. Gas Fee Model : We're implementing Verifier contracts on Ethereum for validating L2 state updates and proofs, enhancing security and trust. Ethereum Layer 1 Verifier : Seamless integration between L1 and Ola L2 will enable users to easily deposit assets from Ethereum to Ola L2. Layer 1 and Layer 2 Deposit and Withdrawal : Our Rollup's DA strategy will support diverse methods, including those offered by Ethereum and Celestia, for robust data availability. Enhanced Data Availability (DA) Layer Integration : Facilitating interoperability and asset transfers between different blockchain networks. Cross-Chain Bridge Functionality : Offering Prover proofs in both CPU and GPU modes, significantly speeding up the proof generation process. GPU-Accelerated Prover : Expanding our support to include advanced cryptographic primitives like EDDSA and BLS signatures. Extended Support for Account Abstraction Cryptographic Primitives : Introducing privacy options to enhance user discretion and security on the platform. Optional Privacy Features Step on the journey with Ola: Empower Individuals with Data Ownership Since November, Ola has received over 200 high-quality dApp deployment applications through our testnet whitelist activity, covering areas such as gaming, SocialFi, and DID. We are excited to witness these projects thrive within Ola's ecosystem. Looking ahead, Ola is dedicated to continuously enhancing testnet, integrating ZK proof privacy with blockchain technology, and setting the stage for our mainnet launch mid-year. We're excited to introduce a new mining mechanism (such as PoW, ZKP, Verifier etc.) and unfold our token economic model in the coming months. This will include detailed incentive rules tailored for our diverse community, including users, developers, miners, and other key stakeholders, marking a significant stride in our journey towards blockchain innovation. Join us in creating a fairer, more decentralized world with Ola. Play your part in shaping the future of Ola, and share in the rewards of our collective success. As we push the boundaries of innovation, we look forward to the remarkable milestones ahead. Thank you for being a vital part of Ola's journey. Let's seize this pivotal moment and embark on a memorable journey towards the Ola Mainnet! About us Ola is an open-source Hybrid ZK Rollup that delivers programmable scalability and privacy to blockchains, empowering individuals through data ownership. Incubated by cryptographers and engineers of Sin7y Labs in 2022, Ola offers a layer2 ZK Stack, featuring Ola-lang, OlaVM, OlaUI, OlaOS, and a series of developer-friendly tools, facilitating the development of any type of contracts. In 2023, Ola launched the Ambassador Program and OVP Points Rewards, now with 20+ global ambassadors. For 2024, Ola plans to roll out public testnet, mainnet, mining mechanism (PoW, ZKP, & Verifier), and a range of incentive programs steadily. If you are interested in joining or partnering with us to contribute to a fair, decentralized, and permissionless world, please contact us at . contact@olavm.org | | | | | | | | | | Website Twitter Discord Telegram Whitepaper GitHub LinkedIn YouTube HackMD Medium HackerNoon

Introducing Ola's Pre-Alpha Testnet: Empowering Data Ownership and ZK Smart Contract Innovation  

Ola-এর প্রি-আলফা টেস্টনেট চালু করা হচ্ছে: ডেটার মালিকানা এবং ZK স্মার্ট চুক্তি উদ্ভাবনের ক্ষমতায়ন

Wir stellen Olas Pre-Alpha-Testnetz vor: Stärkung des Dateneigentums und ZK-Smart-Contract-Innovation

Presentamos la Testnet Pre-Alpha de Ola: potenciando la propiedad de datos y la innovación de contratos inteligentes de ZK

Présentation du testnet pré-alpha d'Ola : renforcer la propriété des données et l'innovation des contrats intelligents ZK

ओला के प्री-अल्फा टेस्टनेट का परिचय: डेटा स्वामित्व और जेडके स्मार्ट कॉन्ट्रैक्ट इनोवेशन को सशक्त बनाना

Ola のプレアルファ テストネットの紹介: データ所有権の強化と ZK スマート コントラクト イノベーション

Ola의 사전 알파 테스트넷 소개: 데이터 소유권 강화 및 ZK 스마트 계약 혁신

Apresentando a rede de teste pré-alfa da Ola: capacitando a propriedade de dados e a inovação de contratos inteligentes ZK

Представляем пре-альфа-тестовую сеть Ola: расширение возможностей владения данными и инновации в смарт-контрактах ZK

Ola'nın Alfa Öncesi Test Ağıyla Tanışın: Veri Sahipliğini Güçlendirme ve ZK Akıllı Sözleşme Yeniliği

Giới thiệu Mạng thử nghiệm Pre-Alpha của Ola: Trao quyền sở hữu dữ liệu và đổi mới hợp đồng thông minh ZK

介绍 Ola 的 Pre-Alpha 测试网：赋能数据所有权和 ZK 智能合约创新

Introduction JavaScript is a single-threaded, popular, and commonly used programming language. So, besides coding in JavaScript, we should know how the code runs internally. In this article, we will be looking through the steps and understand the internal code execution in JavaScript. So, let's dive into it!! What is Execution Context? Execution context is a fundamental topic in JavaScript which refers to the scope or environment in which the JavaScript code is evaluated and executed. Whenever any code runs in JavaScript, it runs inside the execution context. Its Importance Understanding the execution context is important as it helps us to understand the variable accessibility(i.e., the scope of the variables in the code), the value of 'this' keyword which is often misunderstood, and the concept of Hoisting. Types of Execution Context There are three types of execution context: Global Execution Context
Function Execution Context
Eval Function Execution Context Now, let's dive into the main content and look into the creation of execution context. Creation of Execution Context When a code is run in JavaScript, at first, a global execution context is created. The global execution context is allocated to 'this' variable. After that, the creation of the execution context consists of two phases: The Creation phase and the Code Execution Phase. The JS code runs in two phases Not clear enough? Let's understand these phases and the whole process of code execution through an example and diagrams. Let's have a look at a simple code let num1 = 50;
let num2 = 100;
function product(val1 , val2){
    let ans= val1*val2;
    return ans;
}
let result= product(num1 , num2); While running the code, at first, a global execution context is created which is referred to this variable. In the browser, the value of this is the "window" object. All the further execution contexts are created inside this Global execution context. As JavaScript is a single-threaded language, everything inside it works in a definite process. Now, after the creation of the Global execution context, the code execution takes place. In JS, the code execution takes place in two steps, the first is the memory creation phase in which the variables are declared and allocated memory. The second one is the code execution phase in which the variables are assigned their value and the code is run line by line. Memory Creation Phase In this phase, declaration and memory allocation of variables take place within the execution context but are not initialized or assigned any value. In the above example, in the memory creation phase, at first, variables num1 and then num2 are declared which is initialized with the value undefined. After that, the function product is declared which contains the function definition, and lastly, the variable result is declared which contains the value undefined. Here comes the concept of "Hoisting." Hoisting is a JS mechanism where variables and function declarations are moved to the top of their containing scope in the compilation phase even before the execution of the code. Thus, these variables are hoisted to the top of their scopes. Now, let's look into the next phase, i.e., the Execution phase. Execution Phase After the variables are declared, the first half of execution is completed. Now, here starts the code execution phase. In the execution phase, all the declared variables are initialized with the values, and the code is executed line by line. In the above example, during the execution phase at first num1 is initialized with the value, i.e., 50, and similarly, num2 is initialized with 100. Following that the function product is called and for that function, a new function execution context is created consisting of a new variable environment and execution thread. (Refer to the following image for better understanding). The function is also executed similarly in two phases inside the function execution context until it returns the value to the Global execution context. Function Execution Context: In the memory creation phase val1, val2and ans are declared consecutively and are initialized with undefined. After that, the execution phase val1 is initialized with the value 50 as val1= num1. Similarly, val2= 100 as val2=num2. In the next step, ans is evaluated as ans= val1*val2 and assigned with the value 5000. After it returns the value to the global execution context, the function execution context is deleted. In the next step, the returned value by the function is assigned to the result variable, and the output is given. Now, the execution of the code is completed. After completion, the whole execution context is destroyed. There is also a concept called Call Stack. It is a mechanism that keeps track of the functions being called, which are being executed, and which are in queue to be executed one after another. It follows the LIFO principle (i.e., Last In First Out principle). The function called at the last is added to the top of the list and after its execution, it is popped out. Conclusion That was all about how code is executed in JavaScript. I hope I was able to clear the process of code execution more simply. If you found this blog useful and insightful, share it and comment down your views on it. Do follow for more such content. Here's how you can connect with me. Email - olibhiag@gmail.com Socials: Twitter - LinkedIn - GitHub Thanks for giving it a read!! Introduction JavaScript is a single-threaded, popular, and commonly used programming language. So, besides coding in JavaScript, we should know how the code runs internally. In this article, we will be looking through the steps and understand the internal code execution in JavaScript. So, let's dive into it!! What is Execution Context? Execution context is a fundamental topic in JavaScript which refers to the scope or environment in which the JavaScript code is evaluated and executed. Whenever any code runs in JavaScript, it runs inside the execution context. Its Importance Understanding the execution context is important as it helps us to understand the variable accessibility(i.e., the scope of the variables in the code), the value of 'this' keyword which is often misunderstood, and the concept of Hoisting. 'this' 'this' Types of Execution Context There are three types of execution context: Global Execution Context Function Execution Context Eval Function Execution Context Global Execution Context Function Execution Context Eval Function Execution Context Now, let's dive into the main content and look into the creation of execution context. Creation of Execution Context When a code is run in JavaScript, at first, a global execution context is created. The global execution context is allocated to 'this' variable. After that, the creation of the execution context consists of two phases: The Creation phase and the Code Execution Phase. The JS code runs in two phases The Creation phase Code Execution Phase. Not clear enough? Let's understand these phases and the whole process of code execution through an example and diagrams. Let's have a look at a simple code let num1 = 50;
let num2 = 100;
function product(val1 , val2){
    let ans= val1*val2;
    return ans;
}
let result= product(num1 , num2); let num1 = 50;
let num2 = 100;
function product(val1 , val2){
    let ans= val1*val2;
    return ans;
}
let result= product(num1 , num2); While running the code, at first, a global execution context is created which is referred to this variable. this In the browser, the value of this is the "window" object. All the further execution contexts are created inside this Global execution context. As JavaScript is a single-threaded language , everything inside it works in a definite process. In the browser, the value of this is the "window" object. single-threaded language Now, after the creation of the Global execution context, the code execution takes place. In JS, the code execution takes place in two steps, the first is the memory creation phase in which the variables are declared and allocated memory. The second one is the code execution phase in which the variables are assigned their value and the code is run line by line. Memory Creation Phase In this phase, declaration and memory allocation of variables take place within the execution context but are not initialized or assigned any value. In the above example, in the memory creation phase, at first, variables num1 and then num2 are declared which is initialized with the value undefined . After that, the function product is declared which contains the function definition, and lastly, the variable result is declared which contains the value undefined . num1 num2 undefined product result undefined Here comes the concept of "Hoisting." "Hoisting." Hoisting is a JS mechanism where variables and function declarations are moved to the top of their containing scope in the compilation phase even before the execution of the code. Thus, these variables are hoisted to the top of their scopes. Now, let's look into the next phase, i.e., the Execution phase. Execution Phase After the variables are declared, the first half of execution is completed. Now, here starts the code execution phase. In the execution phase, all the declared variables are initialized with the values, and the code is executed line by line. In the above example, during the execution phase at first num1 is initialized with the value, i.e., 50 , and similarly, num2 is initialized with 100 . num1 50 num2 100 Following that the function product is called and for that function, a new function execution context is created consisting of a new variable environment and execution thread. product (Refer to the following image for better understanding). The function is also executed similarly in two phases inside the function execution context until it returns the value to the Global execution context. Function Execution Context : In the memory creation phase val1 , val2 and ans are declared consecutively and are initialized with undefined . After that, the execution phase val1 is initialized with the value 50 as val1= num1 . Similarly, val2= 100 as val2=num2 . In the next step, ans is evaluated as ans= val1*val2 and assigned with the value 5000 . Function Execution Context Function Execution Context val1 val2 ans undefined val1 50 val1= num1 val2= 100 val2=num2 ans= val1*val2 5000 After it returns the value to the global execution context, the function execution context is deleted. In the next step, the returned value by the function is assigned to the result variable, and the output is given. Now, the execution of the code is completed. After completion, the whole execution context is destroyed. result There is also a concept called Call Stack. It is a mechanism that keeps track of the functions being called, which are being executed, and which are in queue to be executed one after another. Call Stack. It follows the LIFO principle (i.e., Last In First Out principle). The function called at the last is added to the top of the list and after its execution, it is popped out. Conclusion That was all about how code is executed in JavaScript. I hope I was able to clear the process of code execution more simply. If you found this blog useful and insightful, share it and comment down your views on it. Do follow for more such content. Here's how you can connect with me. Email - olibhiag@gmail.com olibhiag@gmail.com Socials: Twitter - LinkedIn - GitHub Twitter LinkedIn GitHub Thanks for giving it a read!!

Everything You Need to Know About JavaScript Execution Context

Reinventing Blockchain Security: Exclusive Interview With Cyvers CEO Deddy Lavid

How to Start a Career in Product Marketing: 10 Expert Tips (2024)

The Growing Impact of Electric Vehicles in India's Sustainable Future

Auto-Increment Columns in Databases: A Simple Trick That Makes a Big Difference

Mastering Maps in Go: Everything You Need to Know

গো-তে মানচিত্র আয়ত্ত করা: আপনার যা জানা দরকার

Maps in Go meistern: Alles, was Sie wissen müssen

Dominar mapas en Go: todo lo que necesitas saber

Maîtriser les cartes en Go : tout ce que vous devez savoir

गो में मानचित्रों पर महारत हासिल करना: वह सब कुछ जो आपको जानना चाहिए

Go でマップをマスターする: 知っておくべきことすべて

Go로 지도 마스터하기: 알아야 할 모든 것

Dominando mapas em Go: tudo o que você precisa saber

Освоение карт в Go: все, что вам нужно знать

Go'da Haritalarda Uzmanlaşmak: Bilmeniz Gereken Her Şey

Làm chủ bản đồ trong Go: Mọi thứ bạn cần biết

掌握 Go 中的 Maps：你需要知道的一切

As an early adopter and a crypto enthusiast, I've been closely following the rise of AI Crypto Coins. The year 2024 has been nothing short of a roller coaster ride for us. Ever since OpenAI's ChatGPT was released in late 2022, AI has become all the hype. This mania has since perpetuated to AI crypto coins, with prices surging across the board. Top AI Crypto Coins by Market Cap The top AI crypto coin by market cap is Internet Computer (DFINITY), at $8.28 billion as of April 08, 2024. This is followed by the NEAR Protocol (NEAR), with a market cap of $4.44 billion. Bittensor (TAO) is third, with a market of $3.81 billion. Bittensor, a decentralized machine learning protocol powered by the TAO token, remained the largest AI crypto coin throughout January and February 2024. Movers and Shakers However, outside of the top 5, there were significant changes. Autonolas (OLAS), which started 2024 in sixth place with a market cap of $0.30 billion, dropped to 24th as of April 8, at 172.30 million. Nosana (NOS) did really well in getting higher in the rankings. At the beginning of 2024, it was in ninth place. But by the end of February, it moved up to seventh place. Its market cap, which is how much the company is worth, went up a lot, too. It went from being worth $0.05 billion to $0.51 billion, which is an increase of $0.46 billion in February. But now, it's dropped again to 16th place, and its market cap is $257.21 million. 2024's Biggest Gainers The largest gainer amongst crypto AI coins in 2024 was Nosana, with a price increase of 987.9%, rising from $0.56 to $6.01 within two months. This was followed by DeepFakeAI (FAKEAI), which saw an increase of 628.6%, from $0.0008 to $0.0069. Synesis One (SNS) came in third, rising 595.4%, increasing its price from $0.016 at the start of the year to $0.108 by the end of February. The Impact of OpenAI's Sora The announcement of OpenAI's Sora text-to-video model saw an average price increase of 151.0% amongst AI crypto coins. DeepFakeAI benefited the most from OpenAI's latest product, having managed to reverse its downtrend of -31.4% on February 17, to 628.6% on February 29. Leaders in Market Cap Growth In terms of increase in market cap, Bittensor led the pack, having added $2.22 billion in valuation. It was followed by Render, which increased by $1.20 billion. SingularityNET (AGIX) came in third, with its market cap rising by $0.51 billion since the start of 2024 to $0.93 billion. Looking Ahead: The Future of AI and Crypto As we move forward, the AI-Crypto narrative continues to evolve. The convergence of AI and blockchain technologies is creating new opportunities and challenges. But one thing is certain: the future of AI and Crypto is bright, and I am excited to be part of this journey. Sources: Coingecko, Bitcoininsider, and Bitcoinethereumnews. As an early adopter and a crypto enthusiast, I've been closely following the rise of AI Crypto Coins . The year 2024 has been nothing short of a roller coaster ride for us. Ever since OpenAI's ChatGPT was released in late 2022, AI has become all the hype. This mania has since perpetuated to AI crypto coins, with prices surging across the board. AI Crypto Coins Top AI Crypto Coins by Market Cap The top AI crypto coin by market cap is Internet Computer (DFINITY), at $8.28 billion as of April 08, 2024. This is followed by the NEAR Protocol (NEAR), with a market cap of $4.44 billion . Bittensor (TAO) is third, with a market of $3.81 billion . Bittensor, a decentralized machine learning protocol powered by the TAO token, remained the largest AI crypto coin throughout January and February 2024. $8.28 billion $4.44 billion $3.81 billion Movers and Shakers However, outside of the top 5, there were significant changes. Autonolas (OLAS), which started 2024 in sixth place with a market cap of $0.30 billion , dropped to 24th as of April 8, at 172.30 million . Nosana (NOS) did really well in getting higher in the rankings. At the beginning of 2024, it was in ninth place. But by the end of February, it moved up to seventh place. Its market cap, which is how much the company is worth, went up a lot, too. It went from being worth $0.05 billion to $0.51 billion, which is an increase of $0.46 billion in February. But now, it's dropped again to 16th place, and its market cap is $257.21 million. $0.30 billion 172.30 million 2024's Biggest Gainers The largest gainer amongst crypto AI coins in 2024 was Nosana, with a price increase of 987.9%, rising from $0.56 to $6.01 within two months. This was followed by DeepFakeAI (FAKEAI) , which saw an increase of 628.6%, from $0.0008 to $0.0069 . Synesis One (SNS) came in third, rising 595.4%, increasing its price from $0.016 at the start of the year to $0.108 by the end of February. $0.56 $6.01 (FAKEAI) $0.0008 $0.0069 $0.016 $0.108 The Impact of OpenAI's Sora The announcement of OpenAI's Sora text-to-video model saw an average price increase of 151.0% amongst AI crypto coins. DeepFakeAI benefited the most from OpenAI's latest product, having managed to reverse its downtrend of -31.4% on February 17, to 628.6% on February 29. OpenAI's Sora text-to-video model Leaders in Market Cap Growth In terms of increase in market cap, Bittensor led the pack, having added $2.22 billion in valuation. It was followed by Render, which increased by $1.20 billion . SingularityNET (AGIX) came in third, with its market cap rising by $0.51 billion since the start of 2024 to $0.93 billion . $2.22 billion $1.20 billion $0.51 billion $0.93 billion Looking Ahead: The Future of AI and Crypto As we move forward, the AI-Crypto narrative continues to evolve. The convergence of AI and blockchain technologies is creating new opportunities and challenges. But one thing is certain: the future of AI and Crypto is bright, and I am excited to be part of this journey. Sources: Coingecko, Bitcoininsider, and Bitcoinethereumnews. Sources: Coingecko , Bitcoininsider , and Bitcoinethereumnews . Sources: Coingecko Bitcoininsider Bitcoinethereumnews

Forget Elon Musk, AI Crypto is the REAL Money Maker

This paper is under CC 4.0 license. available on arxiv Authors: (1) Kota Chin, University of Tsukuba, National Institute of Information and Communications Technology Japan; (2) Keita Emura, Kanazawa University, Japan National Institute of Information and Communications Technology Japan; (3) Kazumasa Omote, University of Tsukuba National Institute of Information and Communications Technology Japan. Table of Links Abstract & Introduction Preliminaries Proposed Anonymous Yet Accountable Contract Wallet System Implementation Conclusion & References V. CONCLUSION In this paper, we proposed an anonymous yet accountable contract wallet system based on account abstraction and accountable ring signatures. The proposed system is implemented using Solidity for zkSync. Moreover, we discussed potential of the proposed system, e.g., medical information sharing and asset management. Since the current implementation results using Solidity show the required costs are expensive, our result here might be regarded as somewhat conceptual. However, to the best of our knowledge, no previous implementation result is known that confirms the cost to run an accountable ring signature scheme in Solidity to date, and we believe that our result can be seen as an important stepping stone to provide anonymity and accountability simultaneously in blockchain systems. Investigating other applications of the proposed system will be left to future work. The underlying account ring signature scheme does not provide post-quantum security due to the discrete logarithm-based construction. Thus, it is difficult to accept the current construction as a platform to manage large amounts of assets due to the progress of quantum computing. Because a post-quantum accountable ring signature scheme has been proposed in [7], it would be interesting to employ the scheme, precisely, how to implement it using Solidity is left to future work. The authors would like to thank Dr. Miyako Ohkubo (NICT) for her invaluable comments and suggestions. This work was supported by JSPS KAKENHI Grant Numbers JP21K11897 and JP22H03588. Acknowledgment: REFERENCES [1] Arrest of suspected developer of Tornado Cash. https://www.fiod.nl/arrest-of-suspected-developer-of-tornado-cash/. August 12, 2022. [2] StarkNet. https://starkware.co/starknet/. [3] zkSync. https://zksync.io/. [4] Jean-Philippe Aumasson, Daniel J. Bernstein, Ward Beullens, Christoph Dobraunig, Maria Eichlseder, Scott Fluhrer, Stefan-Lukas Gazdag, Andreas H ¨ulsing, Panos Kampanakis, Stefan K ¨olbl, Tanja Lange, Martin M. Lauridsen, Florian Mendel, Ruben Niederhagen, Christian Rechberger, Joost Rijneveld, Peter Schwabe, and Bas Westerbaan. SPHINCS+: Submission to the NIST post-quantum project, v.3.1. https://sphincs.org/data/sphincs+-r3.1-specification.pdf. [5] Asaph Azaria, Ariel Ekblaw, Thiago Vieira, and Andrew Lippman. MedRec: Using blockchain for medical data access and permission management. In 2016 2nd International Conference on Open and Big Data (OBD), pages 25–30, 2016. [6] Eli Ben-Sasson, Iddo Bentov, Yinon Horesh, and Michael Riabzev. Scalable, transparent, and post-quantum secure computational integrity. IACR Cryptol. ePrint Arch., page 46, 2018. [7] Ward Beullens, Samuel Dobson, Shuichi Katsumata, Yi-Fu Lai, and Federico Pintore. Group signatures and more from isogenies and lattices: Generic, simple, and efficient. In EUROCRYPT, pages 95–126, 2022. [8] Ward Beullens, Shuichi Katsumata, and Federico Pintore. Calamari and Falafl: Logarithmic (linkable) ring signatures from isogenies and lattices. In ASIACRYPT, pages 464–492, 2020. [9] Dmytro Bogatov, Angelo De Caro, Kaoutar Elkhiyaoui, and Bj ¨orn Tackmann. Anonymous transactions with revocation and auditing in hyperledger fabric. In CANS, pages 435–459, 2021. [10] Jonathan Bootle, Andrea Cerulli, Pyrros Chaidos, Essam Ghadafi, Jens Groth, and Christophe Petit. Short accountable ring signatures based on DDH. In ESORICS, pages 243–265, 2015. [11] Vitalik Buterin, Ansgar Dietrichs, Matt Garnett, Will Villanueva, and Sam Wilson. EIP-2938: Account Abstraction. https://eips.ethereum.org/EIPS/eip-2938, 2020. [12] Jan Camenisch, Manu Drijvers, and Maria Dubovitskaya. Practical ucsecure delegatable credentials with attributes and their application to blockchain. In ACM CCS, pages 683–699, 2017. [13] David Chaum and Eug`ene van Heyst. Group signatures. In EUROCRYPT, pages 257–265, 1991. [14] Aisling Connolly, J´er ˆome Deschamps, Pascal Lafourcade, and Octavio Perez-Kempner. Protego: Efficient, revocable and auditable anonymous credentials with applications to hyperledger fabric. In INDOCRYPT, pages 249–271, 2022. [15] L´eo Ducas, Eike Kiltz, Tancr`ede Lepoint, Vadim Lyubashevsky, Peter Schwabe, Gregor Seiler, and Damien Stehl´e. CRYSTALS-Dilithium: A lattice-based digital signature scheme. IACR Transactions on Cryptographic Hardware and Embedded Systems, 2018(1):238–268, 2018. [16] Kai Fan, Shangyang Wang, Yanhui Ren, Hui Li, and Yintang Yang. MedBlock: Efficient and secure medical data sharing via blockchain. Journal of Medical Systems, 42(8):136:1–136:11, 2018. [17] Tomilayo Fatokun, Avishek Nag, and Sachin Sharma. Towards a blockchain assisted patient owned system for electronic health records. Electronics, 10(5), 2021. [18] Pierre-Alain Fouque, Jeffrey Hoffstein, Paul Kirchner, Vadim Lyubashevsky, Thomas Pornin, Thomas Prest, Thomas Ricosset, Gregor Seiler, William Whyte, and Zhenfei Zhang. FALCON: Fast-fourier lattice-based compact signatures over NTRU. https://falcon-sign.info/falcon.pdf. [19] Tomoki Fujitani, Keita Emura, and Kazumasa Omote. A privacypreserving enforced bill collection system using smart contracts. In AsiaJCIS, pages 51–60. IEEE, 2021. [20] Merlin George and Anu Mary Chacko. MediTrans - patient-centric interoperability through blockchain. International Journal of Network Management, 32(3), 2022. [21] Jens Groth. Short non-interactive zero-knowledge proofs. In ASIACRYPT, pages 341–358, 2010. [22] Yiheng Liang. Identity verification and management of electronic health records with blockchain technology. In ICHI, pages 1–3. IEEE, 2019. [23] Joseph K. Liu, Victor K. Wei, and Duncan S. Wong. Linkable spontaneous anonymous group signature for ad hoc groups (extended abstract). In ACISP, pages 325–335, 2004. [24] Xingye Lu, Man Ho Au, and Zhenfei Zhang. Raptor: A practical latticebased (linkable) ring signature. In ACNS, pages 110–130, 2019. [25] Mohammad Moussa Madine, Ammar Ayman Battah, Ibrar Yaqoob, Khaled Salah, Raja Jayaraman, Yousof Al-Hammadi, Sasa Pesic, and Samer Ellahham. Blockchain for giving patients control over their medical records. IEEE Access, 8:193102–193115, 2020. [26] Abdullah Al Omar, Md. Zakirul Alam Bhuiyan, Anirban Basu, Shinsaku Kiyomoto, and Mohammad Shahriar Rahman. Privacy-friendly platform for healthcare data in cloud based on blockchain environment. Future Gener. Comput. Syst., 95:511–521, 2019. [27] Bryan Parno, Jon Howell, Craig Gentry, and Mariana Raykova. Pinocchio: nearly practical verifiable computation. Communications of the ACM, 59(2):103–112, 2016. [28] Sen Qiao, Varun Madathil, and Kemafor Anyanwu. Integrating group signatures in complex decentralized marketplace transactions for improved buyer privacy. In Blockchain, pages 139–148. IEEE, 2022. [29] Ronald L. Rivest, Adi Shamir, and Yael Tauman. How to leak a secret. In ASIACRYPT, pages 552–565, 2001. [30] Alex Roehrs, Cristiano Andre da Costa, and Rodrigo da Rosa Righi. OmniPHR: A distributed architecture model to integrate personal health records. Journal of Biomedical Informatics, 71:70–81, 2017. [31] Teppei Sato, Keita Emura, Tomoki Fujitani, and Kazumasa Omote. An anonymous trust-marking scheme on blockchain systems. IEEE Access, 9:108772–108781, 2021. [32] Kazuo Takaragi, Takashi Kubota, Sven Wohlgemuth, Katsuyuki Umezawa, and Hiroki Koyanagi. Secure revocation features in eKYC - privacy protection in central bank digital currency. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 106-A(3), 2023. [33] Qi Xia, Emmanuel Boateng Sifah, Kwame Omono Asamoah, Jianbin Gao, Xiaojiang Du, and Mohsen Guizani. MeDShare: Trust-less medical data sharing among cloud service providers via blockchain. IEEE Access, 5:14757–14767, 2017. [34] Shouhuai Xu and Moti Yung. Accountable ring signatures: A smart card approach. In CARDIS, pages 271–286, 2004.

Anonymous, Accountable Contract Wallet System With Account Abstraction: Conclusion & References

This paper is under CC 4.0 license. available on arxiv Authors: (1) Joel Jang, CarperAI,University of Washington & Allen Institute for AI; (2) Seungone Kim, KAIST AI; (3) Yizhong Wang, University of Washington; (4) Jack Hessel, University of Washington; (5) Luke Zettlemoyer, Aleph Alpha; (6) Hannaneh Hajishirzi, University of Washington & Allen Institute for AI; (7) Yejin Choi, UC San Diego. 5 CONCLUSION Previous work has shown that adapting LLMs with RLHF helps them generate outputs that are preferred by humans over the supervised fine-tuned counterpart. However, recent work has also pointed out that simply training LLMs to abide by the preference of the general may result in ignoring individual preferences and values. In this work, we provide the first steps to tackle this issue by proposing Reinforcement Learning from Personalized Human Feedback as a multi-objective problem so that LLMs can be aligned to follow conflicting preferences. We propose a promising method called P-SOUPS that is able to composite models trained on single objectives on the fly during inference. We also highlight the scalability of P-SOUPS by showing that it scales linearly, instead of exponentially like the MORL baseline, with regards to the number of new preferences, which is required to provide true personalization to individual users. ACKNOWLEDGMENTS Thanks to Minyoung Hwang, Sungdong Kim, Tim Dettmers, Yoonjoo Lee, and Margaret Li for helpful feedback and discussion. REFERENCES Xiang Ao, Xiting Wang, Ling Luo, Ying Qiao, Qing He, and Xing Xie. PENS: A dataset and generic framework for personalized news headline generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pp. 82–92, Online, August 2021. Association for Computational Linguistics. doi: 10.18653/v1/2021.acl-long.7. URL https: //aclanthology.org/2021.acl-long.7. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, T. J. Henighan, Nicholas Joseph, Saurav Kadavath, John Kernion, Tom Conerly, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Tristan Hume, Scott Johnston, Shauna Kravec, Liane Lovitt, Neel Nanda, Catherine Olsson, Dario Amodei, Tom B. Brown, Jack Clark, Sam McCandlish, Christopher Olah, Benjamin Mann, and Jared Kaplan. Training a helpful and harmless assistant with reinforcement learning from human feedback. ArXiv, abs/2204.05862, 2022a. URL https://api.semanticscholar.org/ CorpusID:248118878. Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, et al. Constitutional ai: Harmlessness from ai feedback. arXiv preprint arXiv:2212.08073, 2022b. Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jer´ emy Scheurer, Javier ´ Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, et al. Open problems and fundamental limitations of reinforcement learning from human feedback. arXiv preprint arXiv:2307.15217, 2023. Deborah Cohen, Moonkyung Ryu, Yinlam Chow, Orgad Keller, Ido Greenberg, Avinatan Hassidim, Michael Fink, Yossi Matias, Idan Szpektor, Craig Boutilier, et al. Dynamic planning in openended dialogue using reinforcement learning. arXiv preprint arXiv:2208.02294, 2022. Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, and Leshem Choshen. Cold fusion: Collaborative descent for distributed multitask finetuning. arXiv preprint arXiv:2212.01378, 2022. Yann Dubois, Xuechen Li, Rohan Taori, Tianyi Zhang, Ishaan Gulrajani, Jimmy Ba, Carlos Guestrin, Percy Liang, and Tatsunori B Hashimoto. Alpacafarm: A simulation framework for methods that learn from human feedback. arXiv preprint arXiv:2305.14387, 2023. Shangbin Feng, Chan Young Park, Yuhan Liu, and Yulia Tsvetkov. From pretraining data to language models to downstream tasks: Tracking the trails of political biases leading to unfair nlp models. arXiv preprint arXiv:2305.08283, 2023. Xinyang Geng, Arnav Gudibande, Hao Liu, Eric Wallace, Pieter Abbeel, Sergey Levine, and Dawn Song. Koala: A dialogue model for academic research. Blog post, April, 1, 2023. Conor F Hayes, Roxana Radulescu, Eugenio Bargiacchi, Johan K ˘ allstr ¨ om, Matthew Macfarlane, ¨ Mathieu Reymond, Timothy Verstraeten, Luisa M Zintgraf, Richard Dazeley, Fredrik Heintz, et al. A practical guide to multi-objective reinforcement learning and planning. Autonomous Agents and Multi-Agent Systems, 36(1):26, 2022. Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. LoRA: Low-rank adaptation of large language models. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum? id=nZeVKeeFYf9. Chengsong Huang, Qian Liu, Bill Yuchen Lin, Tianyu Pang, Chao Du, and Min Lin. Lorahub: Efficient cross-task generalization via dynamic lora composition. ArXiv, abs/2307.13269, 2023. URL https://api.semanticscholar.org/CorpusID:260155012. Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Suchin Gururangan, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. Editing models with task arithmetic. ICLR, 2022. Joel Jang, Seungone Kim, Seonghyeon Ye, Doyoung Kim, Lajanugen Logeswaran, Moontae Lee, Kyungjae Lee, and Minjoon Seo. Exploring the benefits of training expert language models over instruction tuning. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett (eds.), Proceedings of the 40th International Conference on Machine Learning, volume 202 of Proceedings of Machine Learning Research, pp. 14702–14729. PMLR, 23–29 Jul 2023. URL https://proceedings.mlr.press/v202/jang23a. html. Sungdong Kim, Sanghwan Bae, Jamin Shin, Soyoung Kang, Donghyun Kwak, Kang Min Yoo, and Minjoon Seo. Aligning large language models through synthetic feedback. arXiv preprint arXiv:2305.13735, 2023. Hannah Rose Kirk, Bertie Vidgen, Paul Rottger, and Scott A Hale. Personalisation within bounds: A ¨ risk taxonomy and policy framework for the alignment of large language models with personalised feedback. arXiv preprint arXiv:2303.05453, 2023. Noriyuki Kojima, Alane Suhr, and Yoav Artzi. Continual learning for grounded instruction generation by observing human following behavior. Transactions of the Association for Computational Linguistics, 9:1303–1319, 2021. Margaret Li, Suchin Gururangan, Tim Dettmers, Mike Lewis, Tim Althoff, Noah A Smith, and Luke Zettlemoyer. Branch-train-merge: Embarrassingly parallel training of expert language models. arXiv preprint arXiv:2208.03306, 2022. Pan Li and Alexander Tuzhilin. Towards controllable and personalized review generation. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 3237–3245, Hong Kong, China, November 2019. Association for Computational Linguistics. doi: 10.18653/v1/D19-1319. URL https://aclanthology.org/D19-1319. Bodhisattwa Prasad Majumder, Shuyang Li, Jianmo Ni, and Julian McAuley. Generating personalized recipes from historical user preferences. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 5976–5982, Hong Kong, China, November 2019. Association for Computational Linguistics. doi: 10.18653/v1/D19-1613. URL https://aclanthology.org/D19-1613. Pierre-Emmanuel Mazare, Samuel Humeau, Martin Raison, and Antoine Bordes. Training mil- ´ lions of personalized dialogue agents. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pp. 2775–2779, Brussels, Belgium, OctoberNovember 2018. Association for Computational Linguistics. doi: 10.18653/v1/D18-1298. URL https://aclanthology.org/D18-1298. Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, et al. Teaching language models to support answers with verified quotes. arXiv preprint arXiv:2203.11147, 2022. Reiichiro Nakano, Jacob Hilton, S. Arun Balaji, Jeff Wu, Ouyang Long, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, and John Schulman. Webgpt: Browser-assisted question-answering with human feedback. ArXiv, abs/2112.09332, 2021a. URL https://api.semanticscholar.org/CorpusID:245329531. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, et al. Webgpt: Browser-assisted question-answering with human feedback. arXiv preprint arXiv:2112.09332, 2021b. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke E. Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Francis Christiano, Jan Leike, and Ryan J. Lowe. Training language models to follow instructions with human feedback. ArXiv, abs/2203.02155, 2022a. URL https://api.semanticscholar.org/ CorpusID:246426909. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35: 27730–27744, 2022b. Baolin Peng, Chunyuan Li, Pengcheng He, Michel Galley, and Jianfeng Gao. Instruction tuning with gpt-4. arXiv preprint arXiv:2304.03277, 2023. Jonas Pfeiffer, Naman Goyal, Xi Victoria Lin, Xian Li, James Cross, Sebastian Riedel, and Mikel Artetxe. Lifting the curse of multilinguality by pre-training modular transformers. arXiv preprint arXiv:2205.06266, 2022. Alexandre Rame, Guillaume Couairon, Mustafa Shukor, Corentin Dancette, Jean-Baptiste Gaya, Laure Soulier, and Matthieu Cord. Rewarded soups: towards pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards. arXiv preprint arXiv:2306.04488, 2023. Alireza Salemi, Sheshera Mysore, Michael Bendersky, and Hamed Zamani. Lamp: When large language models meet personalization. arXiv preprint arXiv:2304.11406, 2023. Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee, Percy Liang, and Tatsunori Hashimoto. Whose opinions do language models reflect? arXiv preprint arXiv:2303.17548, 2023. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017. Umer Siddique, Paul Weng, and Matthieu Zimmer. Learning fair policies in multi-objective (deep) reinforcement learning with average and discounted rewards. In International Conference on Machine Learning, pp. 8905–8915. PMLR, 2020. David Silver, Satinder Singh, Doina Precup, and Richard S Sutton. Reward is enough. Artificial Intelligence, 299:103535, 2021. Prasann Singhal, Tanya Goyal, Jiacheng Xu, and Greg Durrett. A long way to go: Investigating length correlations in rlhf. arXiv preprint arXiv:2310.03716, 2023. Feifan Song, Bowen Yu, Minghao Li, Haiyang Yu, Fei Huang, Yongbin Li, and Houfeng Wang. Preference ranking optimization for human alignment. arXiv preprint arXiv:2306.17492, 2023. Nisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul F Christiano. Learning to summarize with human feedback. Advances in Neural Information Processing Systems, 33:3008–3021, 2020. Alane Suhr and Yoav Artzi. Continual learning for instruction following from realtime feedback. arXiv preprint arXiv:2212.09710, 2022. Richard S Sutton and Andrew G Barto. Reinforcement learning: An introduction. MIT press, 2018. Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, et al. Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239, 2022. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothee´ Lacroix, Baptiste Roziere, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and ` efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023. Peter Vamplew, Richard Dazeley, Cameron Foale, Sally Firmin, and Jane Mummery. Humanaligned artificial intelligence is a multiobjective problem. Ethics and Information Technology, 20:27–40, 2018. Kristof Van Moffaert and Ann Nowe. Multi-objective reinforcement learning using sets of pareto ´ dominating policies. The Journal of Machine Learning Research, 15(1):3483–3512, 2014. Kristof Van Moffaert, Madalina M Drugan, and Ann Nowe. Scalarized multi-objective reinforce- ´ ment learning: Novel design techniques. In 2013 IEEE symposium on adaptive dynamic programming and reinforcement learning (ADPRL), pp. 191–199. IEEE, 2013. Yizhong Wang, Hamish Ivison, Pradeep Dasigi, Jack Hessel, Tushar Khot, Khyathi Raghavi Chandu, David Wadden, Kelsey MacMillan, Noah A Smith, Iz Beltagy, et al. How far can camels go? exploring the state of instruction tuning on open resources. arXiv preprint arXiv:2306.04751, 2023. Mitchell Wortsman, Suchin Gururangan, Shen Li, Ali Farhadi, Ludwig Schmidt, Michael Rabbat, and Ari S Morcos. lo-fi: distributed fine-tuning without communication. arXiv preprint arXiv:2210.11948, 2022a. Mitchell Wortsman, Gabriel Ilharco, Samir Ya Gadre, Rebecca Roelofs, Raphael Gontijo-Lopes, Ari S Morcos, Hongseok Namkoong, Ali Farhadi, Yair Carmon, Simon Kornblith, et al. Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time. In International Conference on Machine Learning, pp. 23965–23998. PMLR, 2022b. Jeff Wu, Long Ouyang, Daniel M Ziegler, Nisan Stiennon, Ryan Lowe, Jan Leike, and Paul Christiano. Recursively summarizing books with human feedback. arXiv preprint arXiv:2109.10862, 2021a. Yuwei Wu, Xuezhe Ma, and Diyi Yang. Personalized response generation via generative split memory network. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 1956–1970, 2021b. Zeqiu Wu, Yushi Hu, Weijia Shi, Nouha Dziri, Alane Suhr, Prithviraj Ammanabrolu, Noah A Smith, Mari Ostendorf, and Hannaneh Hajishirzi. Fine-grained human feedback gives better rewards for language model training. arXiv preprint arXiv:2306.01693, 2023. Jie Xu, Yunsheng Tian, Pingchuan Ma, Daniela Rus, Shinjiro Sueda, and Wojciech Matusik. Prediction-guided multi-objective reinforcement learning for continuous robot control. In International conference on machine learning, pp. 10607–10616. PMLR, 2020. Jing Xu, Arthur Szlam, and Jason Weston. Beyond goldfish memory: Long-term open-domain conversation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 5180–5197, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.acl-long.356. URL https: //aclanthology.org/2022.acl-long.356. Runzhe Yang, Xingyuan Sun, and Karthik Narasimhan. A generalized algorithm for multi-objective reinforcement learning and policy adaptation. Advances in neural information processing systems, 32, 2019. Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. Personalizing dialogue agents: I have a dog, do you have pets too? arXiv preprint arXiv:1801.07243, 2018. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, et al. Judging llm-as-a-judge with mt-bench and chatbot arena. arXiv preprint arXiv:2306.05685, 2023. Yinhe Zheng, Guanyi Chen, Minlie Huang, Song Liu, and Xuan Zhu. Personalized dialogue generation with diversified traits. arXiv preprint arXiv:1901.09672, 2019. Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving. Fine-tuning language models from human preferences. arXiv preprint arXiv:1909.08593, 2019. A DETAILS OF EVALUATION SETUP We use a modified version of the GPT4 annotation prompt used by Dubois et al. (2023). We modify the criteria to perform the pairwise evaluation from general to a single preference dimension. We also provide 3 demonstrations: one scenario where there is a tie because both responses do not contain any notion of the preference (e.g. both responses do not show any signs of friendliness), one scenario where there is a clear winner, and one scenario where they are both good, but one is better than the other. Simulated pairwise evaluation We recruited 24 crowd workers for our human evaluation. Figure 5 shows the interface used for human evaluation. We consider both the ‘Tie’ and ‘Both are bad’ options to be Ties. B CRITERIA-WISE EVALUATION The criteria-wise win rate (%) across all the methods are shown in Figure 6. The criteria-wise win rate was calculated by getting the average win-rate of the preference combinations that contained the specific preference dimension. For example, when calculating the criteria-wise win rate of ‘Elementary’, we got the average win-rate of the preference combinations that contained the ‘Elementary’ preference, which includes AAA, AAB, ABA, and ABB. C THE FULL LIST OF EVALUATION PROMPTS The full list of evaluation prompts used in our experiments are provided in Table 5 D DETAILED RESULTS FOR HUMAN EVALUATION AND GPT-4 EVALUATION We provide detailed results (win / loss / tie for each of the preference combinations for our main experimental results. Table 6 shows the GPT4 evaluation and Table 7 shows the human evaluation results. E EXAMPLES OF P-SOUPS TEXT GENERATIONS Table 8 shows empirical examples of the text generated from each preference combination of the 16 preference combination experiments for the same prompt.

Personalized Soups: LLM Alignment Via Parameter Merging - Conclusion & References

Well, keyboards are an essential part of daily routine nowadays, especially for programmers & designers. But…In this collection, we are NOT going to discuss any regular keyboards. Instead, you will see some of the strangest, weirdest, unbelievably interesting keyboards that you may hardly believe, can exist..!! So, get ready and explore the collection of the Strangest, weirdest Keyboards. NOTE: Our intention here is just to explore various types of keyboards. Understand that, this is not a review post for any of the mentioned keyboards nor do we encourage anyone to purchase these keyboards. One should do research before purchasing the products. Key Copy Paste by Stack Overflow Starting with the weird Keyboard that will interest lots of programmers due to its unique & weird design. Well, Stack overflow needs no introduction as it is a crucial part of developers’ lives. It is like oxygen for programmers. Teased as an , Stack Overflow later revealed that it is a real keyboard for developers who actively use Stack Overflow..!! April Fool joke Since its launch, it has become the talk of the town due to its interestingly weird look & function. Over 11.8k units have been sold to date..!! So, if you are an avid stack overflow fan and want to speed up your workflow, then this is the one for you. Where to purchase from? You can get your unit from the . Drop How To Configure? Check out the . detailed documentation Figma Creator Micro Keyboard Another keyboard in the list of weirdest keyboards is Figma Creator, which is aimed at a specific community of Figma. It really makes you happy if you get exactly what you imagined. Isn't it? Well, Figma has launched the Creator micro keyboard with their partner Worklouder. The idea took place in 2022 and gained so much praise from the whole Figma community as with the minimalist design this keyboard gives Figma designer a much-needed boost with the customized shortcode key functions. They have literally "cmd+g" ed most used shortcodes. What makes it fantastic is, that you can command all your shortcuts, regardless of the software, via 4 discrete mappable layers. It comes with 30 uniquely Figma icons and 15 solid colors so you can create the perfect look and layout for your setup. The Creator Micro is equipped with 12 keys, with each supporting up to four different functions depending on which "layer" you have the peripheral set to. In total, that gives you access to 48 different shortcuts from the macro pad. Features: 30 custom-designed Figma icons 15 solid color Works with Mac, iOS, Win, Linux, Android Works with all software Here are some of the things you can do with the Figma Creator Micro Keyboard: Create custom shortcuts for your most frequently used Figma commands. Build muscle memory faster by having all your shortcuts in one place. Map functions to the dials for easy access to things like font size, shapes, send to back/front, zoom, and much more. Command all your shortcuts, regardless of the software, via 4 discrete mappable layers. Where to purchase from? You can get your unit from the . Worklouder Who should use it? If you're looking for a way to boost your productivity in Figma, the Figma Creator Micro Keyboard is a great option. It's a well-made, versatile keyboard that can help you save time and effort on your design projects. Well, while working on the Figma project, you can consider using the best . Materio Figma UI Kit & Dashboard Builder It is the best with features like Drag & Drop components, Atomic Design, Auto Layout & 500+ Components. You can build with ease. Figma UI Kit admin dashboards Freestyle-pro-keyboard Have you ever felt the urge to split the keyboard into two pieces so you can freely use both parts comfortably..!!? Well, such keyboards do exist. It’s Kinesis Freestyle Pro which comes in two pieces that you will have to connect. Bad joke sorry..!! Actually, this keyboard is much more than the pieces, it’s an expandable keyboard that allows you to expand the keyboard at your comfort. It features a split-keyboard design that allows you to position the keyboard halves wherever feels most comfortable for your shoulders and forearms. The keyboard also has tenting and negative tilt adjustments, so you can further customize the typing angle to your liking. Features: Split-keyboard design for maximum comfort Tenting and negative tilt adjustments Cherry MX Brown mechanical key switches Full on-board programmability Up to 20 inches of separation between the keyboard halves Compatible with Windows and Mac Who should use it? The Kinesis Freestyle Pro is a great option for anyone who is looking for an ergonomic mechanical keyboard that offers a high degree of comfort, performance, and adjustability. However, it is also one of the more expensive keyboards on the market, so it is important to weigh the cost against the benefits before making a decision. Where to purchase? You can get your unit from the . Official Site How To Configure? Check out the . Official Guide Matias Ergo Pro The Matias Ergo Pro is another excellent option for an ergonomic mechanical keyboard, and it stands as a direct competitor to the Kinesis Freestyle Pro you mentioned earlier. It features a split-keyboard design, which means that the keys are separated into two halves that are angled away from each other. This allows you to type with your hands and wrists in a more natural position, which can help prevent pain and discomfort. If you are looking for an ergonomic, mechanical keyboard that is quiet, customizable, and comfortable to use, then the Ergo Pro is a great option to consider. Features: Ergonomic design can help to reduce strain on hands and wrists Mechanical switches provide a satisfying typing experience Quiet Click switches are quieter than traditional mechanical switches Programmable keys can be customized to your needs Palm supports and tilting/tenting legs allow for a comfortable typing position 3 extra USB 2.0 ports for connecting additional devices Where to purchase? You can get your unit from the . Official Site How To Configure? Here is the detailed Guide Moonlander Mark 1 Another Out of the box split keyboard but with a unique design and functions. Moonlander is a Next-gen Ergonomic keyboard & more than 40k units have been sold..!! And why not? It gives a super funky look along with the rest pad, RGB lights, and a seamless experience. The keyboard comes with integrated wrist rests, but you can fold them out of the way, where they cleverly attach magnetically to the underside of the keyboard. (Folding up the wrist rests also makes the keyboard fit nicely in the included carrying case.) And if you want, you can take the wrist rests off entirely. You can even program the cluster keys for enter, space, back, forward, etc. If you want to change the default setup, you can switch the keys as per your needs and requirements as well. Depending on how much you want to customize your layout, you might spend a long time fiddling with ideas in the configurator tool. It gives you a free hand when it comes to your comfort. And apart from all of the above, if you are someone who likes to experiment with lights, then there is a big surprise for you. There are also programmable RGB LEDs under each key, meaning that in addition to trying to figure out what each key does, you might also delve into how each key looks, too..!! Isn’t it awesome? Features: Split ergonomic design Customizable layout Mechanical key switches RGB backlighting Tenting and tilt adjustment Thumb cluster Low-profile keycaps Carrying case Two-year warranty Where to purchase? You can get your unit from the . (*Unavailable at the time this blog was written) Official Site How To Configure? Check the official Configuration Guide OrbiTouch Keyless Keyboard Yes, you heard it right…!! It is a Keyless keyboard. Not only is the design a bit strange but the way it functions is also interesting. It does not require any finger or wrist movement to operate, making it a valuable tool for people with conditions such as carpal tunnel syndrome, arthritis, etc. It uses two domes that users glide their fingers over to select letters, numbers, and symbols. The domes are surrounded by color-coded rings that correspond to different groups of keys. By moving their fingers in different directions and tapping the domes, users can type without having to press down on individual keys. It also includes a built-in mouse function that allows users to control the cursor and click on buttons, without using a separate mouse…!! However, it wasn’t a successful product as it took too long to type due to its unusual functionality. Features: Built-in mouse function Portable and compact Works with Windows, Mac, iOS, and Android devices. Where to purchase? You can get your unit from the . Drop How To Configure? Check the official Configuration Guide CharaChorder This Keyboard Lets People Type So Fast It’s Banned From Typing Competitions Fun Fact: At first, you may get confused seeing its weird design and it’s natural. Because it doesn’t even look like a “Keyboard” at all..!! CharaChorder is a unique keyboard designed to enable significantly faster typing speeds than are possible with traditional keyboards. It achieves this by using a chording technique, where you press multiple keys simultaneously to represent entire words or phrases. As said above, looks quite different from a traditional keyboard. It has a layout of 22 keys, arranged in a grid. Each key can be pressed individually, or in combination with other keys. The CharaChorder software interprets these combinations and translates them into words and phrases. What makes it a hit is the typing speed. Using this unique keyboard, you can write up to 200 words per minute….!! However, CharaChorder is not for everyone. It has a steep learning curve, and it takes time and practice to become proficient. Additionally, the chording technique can be awkward for some people. Features: 3D Motion Detection Uses a chording technique, where you press multiple keys simultaneously to represent entire words or phrases. character entry (like a traditional keyboard) and fluid chord/character entry Two modes: Utilize a drawing tablet, gaming mouse, CAD mouse, or pet a kitten while still retaining access to all hotkeys literally at your fingertips. Where to purchase? You can get your unit from the . Official Site How To Configure? Check the Guide DROP + OLKB PREONIC KEYBOARD MX KIT V3 The DROP + OLKB Preonic Keyboard MX Kit V3 is a compact, ortholinear mechanical keyboard that is designed for efficiency and portability. Unlike traditional staggered keyboards, the Preonic has keys arranged in straight columns and rows, resembling a grid. It is a collaboration between Drop and OLKB, and it is based on the popular Planck keyboard design. The Preonic has an additional row of keys compared to the Planck, which gives you more flexibility in terms of programming and layout. The Preonic features a hot-swappable PCB, which means that you can easily change the switches without soldering. It also has a USB-C port for connectivity and an anodized aluminum case for durability. It also uses advanced firmware like QMK that allows for extensive customization of key functions and layouts. Features: Compact ortholinear layout Hot-swappable PCB USB-C port Anodized aluminum case Programmable with QMK Kailh hotswap sockets MX compatible Where to purchase? You can get your unit from the . Drop How To Configure? Check the Tutorial QWERKYWRITER Typewriter Mechanical Keyboard If you are someone who is bored with traditional keyboards and wants to try something unique, then this one is going to catch your eye. Here, the vintage blends with the latest tech, and it’s a damn interesting thing. As the name suggests, they have pulled off the typewriter design in the Keyboard..!! Quiet interesting… Unmatched in its aesthetic design, the Qwerkywriter easily is one of the most stunning devices you will ever see but don’t let that distract you! The Qwerkywriter is jam-packed with industry-leading features and patented designs you won’t find anywhere else. They also offer a separate which is cool. numkey pad Additionally, Qwerkywriter S can program up to 15 characters including combinations using modifier keys(i.e. CMD + C for COPY). The return bar factory setting is set to ‘ENTER’. Features: Multi-Device Wireless Connectivity Wired USB mode with NKRO support Dual Scroll Knob Encoders Mechanical Switches Integrated Tablet Stand Macro Return Bar New Scratch Resistant All Aluminum Chassis Typewriter-Inspired® Keycaps Compatible with IOS/Android/MacOS/Win 10 & higher* WINDOWS Devices may require a separate Bluetooth Dongle if using it via Bluetooth on older or incompatible hardware. *NOTE: Tech Spec: Compatibility: Android, Chrome OS, Mac, Windows, iOS Connectivity: Bluetooth, Wireless Features: Mechanical Language / Layout: QWERTY, QWERTY UK, QWERTZ Where to purchase? You can get your unit from the . Official Site How To Configure? Check the Official User Manual iClever BK08 Foldable Bluetooth Keyboard With Touchpad Some people like everything foldable. Be it mobile, table, or be it a keyboard. At some point it is valid, you can utilize the space in your comfort. And this uncommon keyboard is the same. The iClever BK08 Folding Keyboard with Touchpad is a portable keyboard that folds up into a pocket-sized square. It has a built-in touchpad, so you can use it with your phone, tablet, or laptop without needing a separate mouse. The keyboard is made of aluminum and has a backlit keyboard, so you can use it in low-light conditions. You no longer need to worry about the table space once you finish the work. Simply fold the keyboard and put it in a drawer…!! It is as simple as that. Features: Folds up into a pocket-sized square Built-in touchpad Backlit keyboard Aluminum construction Long battery life Where to purchase? You can purchase it on Amazon Maltron Dual-Hand Fully Ergonomic Keyboard A quite unusual keyboard you may ever see. The design is unbelievable and weird. They even have left-hand only & right hand only keyboards which are still acceptable as they can serve people with disability to work with specific hands But, they have a Head/Mouth stick keyboard which you can use with the stick/finger or with a pencil in your mouth….!! Again, may be designed keeping specially-abled people in mind. The makers claim that the two-handed Maltron fully ergonomic keyboard fits the shape of hands and the different lengths of fingers to reduce movement and tension. In this way, it can not only prevent the onset of RSI (Repetitive Strain Injury) symptoms but is an aid to RSI relief and even recovery from RSI pain. Where to purchase? You can get your unit from the . Official Site Wolf King Gaming keyboard This one is next level as the keyboard is specifically manufactured for the Wolf King game..!! Like, how far you can go for the gaming? The Wolf King Warrior Black Gaming Keyboard is a 55-key, red, and black colored keyboard designed specifically for FPS gamers. It has a compact, palm-sized design that makes it easy to take with you to LAN parties, and it features a 3-layer silicon membrane that makes it quiet and durable. The keyboard also has anti-splash protection, so you don’t have to worry about spilling drinks on it. It is the best choice for FPS gamers who are looking for a compact, affordable, and durable keyboard. Not suitable for Regular Work Though..!! Features: 55 keys, including 15 additional keys for commonly used FPS game controls Red and black color scheme Compact, palm-sized design 3-layer silicon membrane for quiet typing Anti-splash protection USB interface TiPY One Hand Keyboard Black The TiPY One Hand Keyboard Black is a one-handed keyboard that integrates mouse functionality. It is designed to be used with one hand, which can be helpful for people with limited hand mobility or for those who want to save space on their desks. The keyboard has a split layout, with the keys for the left and right hands on separate sides. The mouse function is controlled by a trackpad located on the right side of the keyboard. Despite it being designed for disabled people, normal people can use it as well in regular use. Features: Split layout with keys for left and right hand on separate sides. One-handed design: Trackpad on the right side for cursor control. Integrated mouse function: Danish, German, English, Finnish, French, Icelandic, Italian, Dutch, Norwegian, Portuguese, Swedish, Spanish. 12 language support: Adjustable brightness for better visibility in low light. Backlight: Find the perfect angle for comfort and ergonomics. Adjustable tilt: Easy and modern connection standard. USB-C connection: Other features: Macro keys for shortcuts and custom commands. Media control keys. Programmable function keys. Where to purchase? You can purchase this here Conclusion: Well, here we have collected some of the most weird & uncommon keyboards that may shock you..!! Hope you find this collection entertaining as well as unique. Keyboards are indeed an essential part of the routine and especially for the IT people it is more than a friend. In many cases, using different types of keyboards and experimenting with them is a hobby. So, if you are someone who likes to try various keyboards, then this collection will help you. In case you are using any of them or any other weird Keyboards then do share it with us. We will include it in our collection.

10+ Out-of-the-box Keyboards That May Blow Your Fingers!! 🤯

To unlock the main character’s theurgy moves in P3Reload, you have to have certain personas unlocked. So, it’s important to keep expanding the compendium. After explaining what they are and how to use them, the game gives you 1 to start with, but 6 more can be unlocked. Here’s how: Jack Brothers Personas you need: Jack o’Lanter, Jack Frost Its ability: “Deals medium Almighty damage to all foes. High chance of inflicting Down.” King and I Personas you need: Black Frost, King Frost Its ability: “Heavy Ice damage to all foes, ignoring resistances. Medium chance of Freeze.” Best Friends Personas you need: Decarabia, Forneus Its ability: “Next physical/magic attack will deal more than double damage for 1 ally.” Scarlet Havoc Personas you need: Siegfried, Mithras Its ability: “Deals severe Slash damage to all foes, ignoring resistances.” Trickster Personas you need: Susano-o, Loki Its ability: “Massive Almighty damage to all foes. High chance of inflicting an ailment.” Armageddon Personas you need: Helel, Satan Its ability: “Ultimate Almighty damage to all foes, but reduce’s user’s HP and SP to 1.” Are the Theurgy Special Moves Necessary? Necessary? No, you can get by without unlocking them or even using them. But if you’re struggling to get through Tartarus, especially the Gatekeepers, the Theurgy moves really come in handy. Regardless of the boss, its health, or how high a level it is, the Theurgy can put a sizable dent into their life bar. On top of the Gatekeepers, they work really well on the Monad bosses. Those bosses aren’t particularly hard but when you have to fight 3 back to back, having some extra firepower doesn’t hurt. And if you’re worried about these Theurgy moves making the experience too easy for you, you can just not use them. That’s the great thing about them. Which Theurgy Special Move Is the Best? The one that I pulled out the most was Scarlet Havoc. It really came in clutch when I wanted to steamroll Gatekeeper after Gatekeeper, and it’s legitimately one of the best if not the best in the game. However, the most powerful by far is Armageddon. It does 9,999 damage to all enemies. Which sounds great, until you realize it essentially wipes out your character’s HP and SP; it completely reduces both to 1. This isn’t really that big of a deal if you have a healer on your team, and you can regain SP if you have an invigorate ability, but it can be risky if you’re careless. So, while the other special moves aren’t bad, for my buck, Scarlet Havoc and Armageddon are the two best Theurgy Special Moves in Persona 3 Reload. Read More: The MCU Phase 4 Movies in Order The MCU Phase 3 Movies in Order The Apollo Justice Trilogy: The 5 Best Cases Feature image source

How to Unlock All the MC's Theurgy Super Moves in Persona 3 Reload

পারসোনা 3 রিলোড-এ সমস্ত MC এর Theurgy সুপার মুভগুলি কীভাবে আনলক করবেন

So schalten Sie alle Theurgy-Supermoves des MC in Persona 3 Reload frei

Cómo desbloquear todos los súper movimientos de teurgia del MC en Persona 3 Reload

Comment débloquer tous les super mouvements de théurgie du MC dans Persona 3 Reload

पर्सोना 3 रीलोड में एमसी के सभी थर्गी सुपर मूव्स को कैसे अनलॉक करें

ペルソナ 3 リロードですべての MC のテウルギー スーパー ムーブのロックを解除する方法

페르소나 3 리로드에서 모든 MC의 Theurgy 슈퍼 무브를 잠금 해제하는 방법

Como desbloquear todos os super movimentos de teurgia do MC no Persona 3 Reload

Как разблокировать все суперприемы теургии MC в Persona 3 Reload

Persona 3 Reload'da MC'nin Tüm Theurgy Süper Hareketlerinin Kilidini Açma

Cách mở khóa tất cả các chiêu thức siêu phàm của MC trong Persona 3 Tải lại

如何解锁《女神异闻录 3：重装上阵》中 MC 的所有神术超级动作

On February 26th, Nigeria’s Department of State Security detained two senior executives of Binance as part of investigations into the crypto exchange’s operations in the country. The two executives had their passports seized and were quickly arrested after being invited to Nigeria to answer questions regarding the company’s operations. The identities of the executives at Binance remain undisclosed at present. However, suggest that the duo comprises citizens of both the United States and the United Kingdom. reports The Central Bank of Nigeria has accused Binance of funneling more than $26 billion worth of untraceable funds out of the African nation. Nigerian officials are reportedly investigating Binance for facilitating terrorism financing, money laundering, and other forms of illicit finance. Nigeria’s inquiry into the crypto exchange has reached a critical juncture, with authorities now insisting on access to all user data and information held by Binance Nigeria. “We are concerned that certain practices go on that indicate illicit flows going through a number of these entities and suspicious flows at best. In the case of Binance, in the last one year alone, $26 billion has passed through Binance Nigeria from sources and users who we cannot adequately identify. We are determined to do everything it takes to ensure that we take charge of our market, or to put it differently, do not allow others to manipulate our markets in a way that ends up distortionary.” - , Governor of the Central Bank of Nigeria Olayemi Cardoso Back in September of 2023, Nigeria’s Securities and Exchange Commission issued a disclaimer against Binance Nigeria Limited, stating that the platform was “neither registered nor regulated by the Commission,” thus deeming its operations illegal. and have reported that the Nigerian Communications Commission has ordered telecommunication companies and internet service providers to block access to Binance, Coinbase, Kraken, and various other crypto platforms. Bloomberg Nigerian news outlets Contrary to the various news reports, Coinbase that Nigerian users still have access to the platform. states Do you think Binance will get kicked out of Nigeria just like they were in the United States? What are your thoughts on this news? Join the conversation below! Check out my Socials: https://linktr.ee/zerorequiem0x Also published . here

Binance Under Fire: Executives Detained in Nigeria Amid Investigations

This paper is available on arxiv under CC 4.0 license. Authors: (1) Michael Günther, michael.guenther; (2) Jackmin Ong, jackmin.ong; (3) Isabelle Mohr, isabelle.mohr; (4) Alaeddine Abdessalem, alaeddine.abdessalem; (5) Tanguy Abel, tanguy.abel; (6) Mohammad Kalim Akram, kalim.akram; (7) Susana Guzman, susana.guzman; (8) Georgios Mastrapas, georgios.mastrapas; (9) Saba Sturua, saba.sturua; (10) Bo Wang, bo.wang; (11) Maximilian Werk, maximilian.werk; (12) Nan Wang, nan.wang; (13) Han Xiao, han.xiao}@jina.ai. Table of Links Abstract & Introduction Related Work Training Process Overview Backbone Pre-training Fine-Tuning for Embeddings Evaluation Conclusion & References Appendix 7 Conclusion We have introduced Jina Embeddings v2, a novel embedding model based on a modified BERT architecture. This model eschews positional embeddings and instead employs bi-directional ALiBi slopes to capture positional information. By training a series of embedding models with this innovative architecture on the Web document corpus C4 and subsequently fine-tuning them, we have enabled the encoding of the semantics of both short and long textual values into meaningful vector representations. This effort has produced a new suite of open-source embedding models capable of encoding texts containing up to 8192 tokens. These embeddings signify a 16x increase in the maximum sequence length compared to leading open-source embedding models. Additionally, our model suite exhibits competitive performance on the MTEB benchmark. We also demonstrate how utilizing extended sequence lengths can offer our models an advantage over those without such capabilities. References Nils Reimers and Iryna Gurevych. Sentence-bert: Sentence embeddings using siamese bert-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Lan-guage Processing (EMNLP-IJCNLP), pages 3982– 3992, 2019. Ofir Press, Noah A. Smith, and Mike Lewis. Train short, test long: Attention with linear biases enables input length extrapolation, 2022. Michael Günther, Louis Milliken, Jonathan Geuter, Georgios Mastrapas, Bo Wang, and Han Xiao. Jina embeddings: A novel set of high-performance sentence embedding models. arXiv preprint arXiv:2307.11224, 2023. Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, and Furu Wei. Text embeddings by weaklysupervised contrastive pre-training. arXiv preprint arXiv:2212.03533, 2022. Scott Deerwester, Susan T. Dumais, George W. Furnas, Thomas K. Landauer, and Richard Harshman. Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41(6):391–407, 1990. David Blei, Andrew Ng, and Michael Jordan. Latent dirichlet allocation. In T. Dietterich, S. Becker, and Z. Ghahramani, editors, Advances in Neural Information Processing Systems, volume 14. MIT Press, 2001. URL https://proceedings. neurips.cc/paper_files/paper/2001/file/ 296472c9542ad4d4788d543508116cbc-Paper. pdf. Tianyu Gao, Xingcheng Yao, and Danqi Chen. Simcse: Simple contrastive learning of sentence embeddings, 2022. Luyu Gao and Jamie Callan. Condenser: a pre-training architecture for dense retrieval, 2021. Shitao Xiao, Zheng Liu, Yingxia Shao, and Zhao Cao. Retromae: Pre-training retrieval-oriented language models via masked auto-encoder, 2022. Zehan Li, Xin Zhang, Yanzhao Zhang, Dingkun Long, Pengjun Xie, and Meishan Zhang. Towards general text embeddings with multi-stage contrastive learning, 2023. Nandan Thakur, Nils Reimers, Andreas Rücklé, Abhishek Srivastava, and Iryna Gurevych. Beir: A heterogenous benchmark for zero-shot evaluation of information retrieval models, 2021. Niklas Muennighoff, Nouamane Tazi, Loïc Magne, and Nils Reimers. Mteb: Massive text embedding benchmark, 2023. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding, 2019. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized bert pretraining approach, 2019a. Yann N. Dauphin, Angela Fan, Michael Auli, and David Grangier. Language modeling with gated convolutional networks. CoRR, abs/1612.08083, 2016. URL http://arxiv.org/abs/1612.08083. Noam Shazeer. GLU variants improve transformer. CoRR, abs/2002.05202, 2020. URL https://arxiv. org/abs/2002.05202. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. CoRR, abs/1706.03762, 2017. URL http: //arxiv.org/abs/1706.03762. Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper, and Bryan Catanzaro. Megatron-lm: Training multi-billion parameter language models using model parallelism. CoRR, abs/1909.08053, 2019. URL http://arxiv.org/ abs/1909.08053. Toan Q. Nguyen and Julian Salazar. Transformers without tears: Improving the normalization of selfattention. CoRR, abs/1910.05895, 2019. URL http://arxiv.org/abs/1910.05895. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1): 5485–5551, 2020. Ilya Loshchilov and Frank Hutter. Fixing weight decay regularization in adam. CoRR, abs/1711.05101, 2017. URL http://arxiv.org/abs/1711.05101. Paulius Micikevicius, Sharan Narang, Jonah Alben, Gregory Diamos, Erich Elsen, David Garcia, Boris Ginsburg, Michael Houston, Oleksii Kuchaiev, Ganesh Venkatesh, et al. Mixed precision training. In International Conference on Learning Representations, 2018. Jeff Rasley, Samyam Rajbhandari, Olatunji Ruwase, and Yuxiong He. Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 3505–3506, 2020. Zhuyun Dai, Vincent Y Zhao, Ji Ma, Yi Luan, Jianmo Ni, Jing Lu, Anton Bakalov, Kelvin Guu, Keith Hall, and Ming-Wei Chang. Promptagator: Few-shot dense retrieval from 8 examples. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id= gmL46YMpu2J. Aäron van den Oord, Yazhe Li, and Oriol Vinyals. Representation learning with contrastive predictive coding. CoRR, abs/1807.03748, 2018. URL http: //arxiv.org/abs/1807.03748. Feng Wang and Huaping Liu. Understanding the behaviour of contrastive loss. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 2495–2504. IEEE, 2021. Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, et al. Ms marco: A human generated machine reading comprehension dataset. arXiv preprint arXiv:1611.09268, 2016. Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Matthew Kelcey, Jacob Devlin, Kenton Lee, Kristina N. Toutanova, Llion Jones, Ming-Wei Chang, Andrew Dai, Jakob Uszkoreit, Quoc Le, and Slav Petrov. Natural questions: a benchmark for question answering research. Transactions of the Association of Computational Linguistics, 2019. Samuel Bowman, Gabor Angeli, Christopher Potts, and Christopher D Manning. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 632–642, 2015. Tianqi Chen, Bing Xu, Chiyuan Zhang, and Carlos Guestrin. Training deep nets with sublinear memory cost. arXiv preprint arXiv:1604.06174, 2016. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized BERT pretraining approach. CoRR, abs/1907.11692, 2019b. URL http://arxiv. org/abs/1907.11692. Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. GLUE: A multi-task benchmark and analysis platform for natural language understanding. CoRR, abs/1804.07461, 2018. URL http://arxiv.org/abs/1804.07461. Jason Phang, Thibault Févry, and Samuel R. Bowman. Sentence encoders on stilts: Supplementary training on intermediate labeled-data tasks. CoRR, abs/1811.01088, 2018. URL http://arxiv.org/ abs/1811.01088. Eva Sharma, Chen Li, and Lu Wang. BIGPATENT: A large-scale dataset for abstractive and coherent summarization. CoRR, abs/1906.03741, 2019. URL http://arxiv.org/abs/1906.03741. Wikimedia Foundation. Wikimedia downloads, 2022. URL https://dumps.wikimedia.org. David Wadden, Shanchuan Lin, Kyle Lo, Lucy Lu Wang, Madeleine van Zuylen, Arman Cohan, and Hannaneh Hajishirzi. Fact or fiction: Verifying scientific claims. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 7534–7550, 2020.

JINA EMBEDDINGS 2: 8192-Token General-Purpose Text Embeddings: Conclusion & References

Introduction Security - the most pressing concern that every single country of the world faces in current times. But the world we live in is not as simple as it used to be back in the day. It is more divided than ever before. With conflicts like the Russia-Ukraine conflict, the Palestinian crisis, and the trade war between two industrial and technological giants of the globe, how can the global community come together to fight the prevailing challenges of the 21st century? Must there be losers in order to have new winners on the global screen? Can we achieve a positive-sum goal in the current situation? And how will the important stakeholders of global security come together in such a radicalized environment? The 60th Munich Security Conference (MSC) held from 16th February to 18th February 2024 tried to address all these questions and much more. Here’s what went on in MSC 2024! Key Speakers It was a star-studded event; if we pretend the world leaders to be stars for a while. Global leaders from around 50 countries gathered to put forward a collective policy to counter the newer challenges to the world. Some of the prominent figures included the following. Antonio Guterres, UN Secretary-General Kamala Harris, US Vice President Jens Stoltenberg, NATO Secretary General Wang Yi, China’s Foreign Minister Antony Blinken, US Secretary of State Ursula von der Leyen, President of the European Commission Volodymyr Zelenskyy, Ukrainian President David Cameroon, UK’s Foreign Secretary Olaf Scholz, German Chancellor Josep Borrell, Foreign Policy Chief of the European Union John Kerry, US President’s Special Envoy on Climate Change Apart from these, Israel’s President Isaac Herzog and Palestinian Prime Minister Muhammad Shtayyeh were also among the key speakers. Agenda The agenda of the Munich Security Conference 2024 included dialogue and discussions on a wide array of topics. These topics are of great importance, especially in the context of recent global and regional developments. The major talking points of the MSC 2024 were as follows. - Supply chain dependencies, nuclear security of Europe, the role of the US in the global world, an international order that serves everyone, protection of humanitarian law, fight against disinformation, water security, corruption, climate finance, global refugee crisis, the age and race of AI, food & economic security, tech governance, and recovery plan for Ukraine were the topics discussed on the opening day of the MSC 2024. Day 1: Global Security Challenges - Cybersecurity in current times, Germany’s position in the current world, Ukraine and transatlantic security going forward, neutrality, a collectively defended Europe, protection of global commons, China in the current global order, climate diplomacy, increase the shared opportunities, potential reforms in the UN Security Council, and advantages of climate diplomacy were debated on in the first half of day 2. Day 2 (Part I): International Order - Collective strategic efforts to counter crises, Indo-Pacific security, the war in Sudan, ensuring Ukraine’s victory against Russia, stabilization of the Sahel, Russia’s future, Red Sea, Arctic defense, Middle East, Palestine, Israel, and stability, Balkan states, and civil resistances in different areas of the world were the highlight topics from second half day 2 of the Munich Security Conference 2024. Day 2 (Part II): Regional Conflicts & Crises - The EU’s partnership with its partners, the possibility of peace between Israel and Palestine, the expansion of the European Union, underscoring the EU’s geopolitical plan, and the future of geopolitics were the topics on the last day’s agenda followed by closing remarks by the Chairman of the Munich Security Conference. Day 3: Europe in the World Current Global Security Trends: A Lose-Lose Situation The end of the Cold War was one of the crucial turning points in history. Countries inclined towards democracy since the Marxism-Leninism ideology of the Soviet Union had failed. The proverbial “pie” increased exponentially all across the globe because of a rule-based world order, ushering in an era of prosperity. This new world order promised to be the solution to all of the problems that the Global South was facing at the time. Undelivered Promises “Without genuine solidarity, there can never be real trust.” (Subrahmanyam Jaishankar - Indian External Affairs Minister) It’s been more than three decades since the world adopted a new order by and large, but the issues of the developing world have only increased. The Global South believes that the new world order could not deliver what it had promised, and the share of the “pie” has remained the same, if not decreased, for the underdeveloped world. The United States is blamed to be creating hindrances for countries like China - which have greatly benefited from the liberal economic order - to achieve their goals. Relative-Gains Thinking Among Global North and South The debate becomes interesting because the Global North is also dissatisfied with its current share of the pie, and many people in the believe that the obvious rise of and other such economies in the coming decade will stifle growth in their own countries. G7 China This never-ending cycle has come into being because of a number of factors, such as rising global conflicts, stagnation in the financial growth of people, and the relative comparison of success among different global stalwarts. The result is a situation that is slowly becoming visible in many global and regional policies. lose-lose Zero-Sum Beliefs in Russia-Ukraine Conflict The relative-gains thinking has resulted in many countries falsely believing that other nations prosper only by using their share of the This has pushed such countries toward autocratic tendencies and a conflictual mindset because they have one goal now: to widen their sphere of influence as much as they can. Russia’s invasion of Ukraine is an apt example in this regard. pie. Vlodymyr Zelenskyy said, “Ukraine has already shattered myths.” The two-year-long war has completely changed the regional dynamics. Although the war has stayed longer than Russia initially thought, Ukraine might completely lose its sovereignty and territory to Russia, while Europe will have to spend extra on its defense besides giving financial support to Ukraine. All is not well in Russia itself either, as the population is facing issues like rising inflation and an increased crackdown on anti-government activists. Indo-Pacific Region: The Next Hotspot! The highly important Indo-Pacific region seems to be the next location of a global conflict with far-reaching consequences. Now, you must be wondering why the region is so important. Well, this is because the Indo-Pacific region sees 60 percent of maritime trade pass through it. With China gradually trying to increase its military presence in regional waters, many East Asian countries are going into the US camp, both financially and security-wise. However, the highly integrated global trade system means everyone, including China itself, will only lose if the Indo-Pacific becomes the region that sees an escalation in the power rivalry. The Unstable Middle East Just when the Middle East region had started showing some stability with many important regional countries starting or restoring diplomatic and economic relations with Israel, Hamas’s attack on Israel last October pushed the era back to unrest. It is pertinent to mention here that civilians are the biggest losers in this conflict, with scores of injuries and casualties in Gaza and the complete destruction of the infrastructure. The fire of this conflict has reached Tehran, and Iran has also become active through its proxies. The EU’s foreign affairs chief, Josep Borrell, expressed his concerns about Israel’s strategy in the ongoing conflict and raised his voice in favor of a two-state solution. He said, "Hamas is an idea, and you don't kill an idea" The Shaky Sahel Recent developments in the African Sahel have meant the return of authoritarian and autocratic regimes in the region. Many countries, including Niger and Mali, have seen the overthrow of democratic governments by military juntas. This has posed a serious problem for the US and Europe, as their democratic agenda for the region has come to a halt. But the local population is facing the lose-lose situation the most. And it was because of this that Bola Tinubu, Nigerian President and Chairman of the Economic Community of West African States (ECOWAS) said last year, “We will not allow coup after coup in West Africa again.” Receding Globalization The world has reaped great rewards from its economic and technological interconnectedness and interdependence. But, mutually beneficial policies are no longer the preference for many countries as self-reliance has come to the fore. The recent World Economic Situation and Prospects report predicts that the global GDP growth is set to decelerate to 2.4% in 2024 from 2.7% in 2023. To highlight the importance of globalization, China’s foreign minister, Wang Yi said, "The world economy is like a big ocean that cannot be cut into isolated lakes” These changing dynamics might suit the rich and powerful countries, but the low-income nations will have to pay a lot if the fragmentation of the global economy continues. Threatened Global Climate Policy Countries are increasingly moving towards serving their national interests at all costs. They are doing so at the cost of global interests, like collective efforts towards curbing the effects of climate change. Global issues like the US-China trade war and disagreements over climate financing are posing a serious threat to global climate policy. The world is heating up, both literally and figuratively. Artificial Intelligence & Cyber Warfare Everyone is in the race to win technological warfare. This has become evident in the increasing cases of cyber warfare. In recent years, there has been an increase in cyber attacks, which have become a serious challenge for many countries. Artificial Intelligence has only made the competition more fierce. According to Forbes, the AI industry is expected to be at the That is a staggering increase from its market size in 2022. $407 billion mark by 2027. $320 billion Moreover, the Tom & Jerry race between China and the US continues. Both countries want to surpass each other in the semiconductor industry. This could result in a zero-sum game for both countries as well as many other countries in the world. This grave situation warrants that artificial intelligence must have some checks and balances put on it before it is too late. Key Takeaways from MSC 2024 - A Global Order That Benefits All The rule-based international order is in desperate need of change. However, the global community has all its efforts focused on policies and moves that are creating further cracks in the liberal economic order. The need of the hour is to get out of the vicious cycle of relative-gain thinking and mold policies in such a way that the end result is the formulation and implementation of positive-sum policies. However, the road ahead is not easy and needs a concentrated global and regional effort to move from a lose-lose situation to a win-win situation for everyone. To conclude the whole debate in the words of Kamala Harris, the U.S. Vice President, `“Isolation is not insulation.”`

Munich Security Conference 2024: Everything That Went Down

Stories, News & Mentions about OLAB

Olab’s Stories on HackerNoon