Qeeper

qeeper

Hey Hackers! As seasoned HackerNoon users, you're probably well-acquainted with tags as a means to discover stories aligned with your interests. Now, we've taken it up a notch! For the first time since we launched HackerNoon 2.0 in 2019, the tag pages, which organize the entirety of the HackerNoon library into niches, have gotten a total revamp! Our newly updated tag index page makes it easier than ever to find exactly what you're looking for. What’s new on tag pages? Explore Over 88,000 Tags on the Tag Index Page With the introduction of three new filter sections, users can now discover popular tags in different ways including “Most Used”, “Trending” or “Last Published”. Additionally, we’ve also included the number of stories and a brief description, giving you a clearer understanding of each one. Users can also explore tags under a “Parent Category” - HackerNoon’s Curation of Essential Tech Categories on the Internet. The rest of the tags are just a few scrolls away, so take your time to explore and make the most of this page! New Individual Tag Pages Individual tag pages have been enhanced to be more engaging and informative than before, with: User Testimonials: Featuring quotes and comments from the HackerNoon community
Writing Contest: Discover ongoing contests on your favorite tags.
Unique Banner Images: Inspired by the iconic HackerNoon font, our design team has artfully crafted each tag's text into a unique pattern, creating a one-of-a-kind banner image that adds a personal touch to users’ browsing experience.
Action Buttons: Options to subscribe to each tag, start writing, or explore related topics. Plus, we’ve included a search bar to help you find sub-tags within the main tag, resulting in an infinite list of stories that match your search queries. You can also filter results by "Most Read" or "Most Recent". This new setup allows you to effortlessly browse all your desired stories without the need to click through multiple pages. How to Search for Tags Step 1: Enter the desired keywords in the search box, located in the upper right corner of the page.


Step 2: As the results are generated, you’ll discover a list of all related tags, along with the number of stories published under each one. Select the tag that sparks your interest—for example, let’s go with #hackernoon-product-update. Step 3: The selected tag will take you to its dedicated page, where you can subscribe for updates (wink wink!). Just below the title and testimonial sections at the top of the page, you’ll notice a search bar. Use this to narrow down your search with secondary tags. For instance, if you type #gif, your results will be filtered down to just a few relevant stories, making it easy to find exactly what you’re looking for. Simple, right? Dive in and start exploring—refining your searches has never been easier. For more tips on how to maximize the value of tags on your stories, check out this guide and this one. Hey Hackers! As seasoned HackerNoon users, you're probably well-acquainted with tags as a means to discover stories aligned with your interests. Now, we've taken it up a notch! For the first time since we launched HackerNoon 2.0 in 2019, the tag pages, which organize the entirety of the HackerNoon library into niches, have gotten a total revamp! For the first time since we launched HackerNoon 2.0 in 2019 , the tag pages , which organize the entirety of the HackerNoon library into niches, have gotten a total revamp! since we launched HackerNoon 2.0 in 2019 tag pages Our newly updated tag index page makes it easier than ever to find exactly what you're looking for. tag index page tag index page What’s new on tag pages? What’s new on tag pages? What’s new on tag pages? Explore Over 88,000 Tags on the Tag Index Page Explore Over 88,000 Tags on the Tag Index Page With the introduction of three new filter sections, users can now discover popular tags in different ways including “Most Used” , “Trending” or “Last Published” . Additionally, we’ve also included the number of stories and a brief description, giving you a clearer understanding of each one. Users can also explore tags under a “Parent Category” - HackerNoon’s Curation of Essential Tech Categories on the Internet. “Most Used” “Trending” “Last Published” “Parent Category” “Parent Category” The rest of the tags are just a few scrolls away, so take your time to explore and make the most of this page! New Individual Tag Pages New Individual Tag Pages New Individual Tag Pages Individual tag pages have been enhanced to be more engaging and informative than before, with: User Testimonials: Featuring quotes and comments from the HackerNoon community Writing Contest: Discover ongoing contests on your favorite tags. Unique Banner Images: Inspired by the iconic HackerNoon font, our design team has artfully crafted each tag's text into a unique pattern, creating a one-of-a-kind banner image that adds a personal touch to users’ browsing experience. Action Buttons: Options to subscribe to each tag, start writing, or explore related topics. User Testimonials: Featuring quotes and comments from the HackerNoon community User Testimonials: Writing Contest: Discover ongoing contests on your favorite tags. Writing Contest: Unique Banner Images : Inspired by the iconic HackerNoon font, our design team has artfully crafted each tag's text into a unique pattern, creating a one-of-a-kind banner image that adds a personal touch to users’ browsing experience. Unique Banner Images Action Buttons : Options to subscribe to each tag, start writing, or explore related topics. Action Buttons Plus, we’ve included a search bar to help you find sub-tags within the main tag, resulting in an infinite list of stories that match your search queries. You can also filter results by "Most Read" or "Most Recent". This new setup allows you to effortlessly browse all your desired stories without the need to click through multiple pages. How to Search for Tags Step 1: Enter the desired keywords in the search box, located in the upper right corner of the page. Step 2: As the results are generated, you’ll discover a list of all related tags, along with the number of stories published under each one. Select the tag that sparks your interest—for example, let’s go with #hackernoon-product-update. Step 1: Enter the desired keywords in the search box, located in the upper right corner of the page. Step 1: Enter the desired keywords in the search box, located in the upper right corner of the page. Step 2: As the results are generated, you’ll discover a list of all related tags, along with the number of stories published under each one. Select the tag that sparks your interest—for example, let’s go with #hackernoon-product-update. Step 2: As the results are generated, you’ll discover a list of all related tags, along with the number of stories published under each one. Select the tag that sparks your interest—for example, let’s go with #hackernoon-product-update . #hackernoon-product-update Step 3: The selected tag will take you to its dedicated page, where you can subscribe for updates (wink wink!). Just below the title and testimonial sections at the top of the page, you’ll notice a search bar. Use this to narrow down your search with secondary tags. For instance, if you type #gif, your results will be filtered down to just a few relevant stories, making it easy to find exactly what you’re looking for. Simple, right? Step 3: The selected tag will take you to its dedicated page, where you can subscribe for updates (wink wink!). Just below the title and testimonial sections at the top of the page, you’ll notice a search bar. Use this to narrow down your search with secondary tags. For instance, if you type #gif, your results will be filtered down to just a few relevant stories, making it easy to find exactly what you’re looking for. Simple, right? Step 3: The selected tag will take you to its dedicated page, where you can subscribe for updates (wink wink!). Just below the title and testimonial sections at the top of the page, you’ll notice a search bar. Use this to narrow down your search with secondary tags. For instance, if you type #gif, your results will be filtered down to just a few relevant stories, making it easy to find exactly what you’re looking for. Simple, right? Dive in and start exploring —refining your searches has never been easier. start exploring start exploring For more tips on how to maximize the value of tags on your stories, check out this guide and this one. For more tips on how to maximize the value of tags on your stories, check out this guide and this one . this guide this guide this one this one

Tagception: Dive Deeper with HackerNoon's Enhanced Tag Search!

ট্যাগসেপশন: হ্যাকারনুন এর উন্নত ট্যাগ অনুসন্ধানের সাথে আরও গভীরে যান!

Tagception: Tauchen Sie tiefer ein mit der erweiterten Tag-Suche von HackerNoon!

Tagception: ¡Sumérjase más profundamente con la búsqueda de etiquetas mejorada de HackerNoon!

Tagception : plongez plus profondément avec la recherche de tags améliorée de HackerNoon !

टैगसेप्शन: हैकरनून की उन्नत टैग खोज के साथ गहराई से गोता लगाएँ!

Tagception: HackerNoon の強化されたタグ検索でさらに深く探ってみましょう!

Tagception: HackerNoon의 강화된 태그 검색으로 더욱 깊이 있게 알아보세요!

Tagception: mergulhe mais fundo com a pesquisa de tags aprimorada do HackerNoon!

Tagception: погружайтесь глубже с расширенным поиском по тегам HackerNoon!

Tagception: HackerNoon'un Gelişmiş Etiket Aramasıyla Daha Derine Dalın!

Tagception: Khám phá sâu hơn với tính năng Tìm kiếm thẻ nâng cao của HackerNoon!

Tagception：利用 HackerNoon 的增强标签搜索进行更深入的探索！

AI technology is firmly carving out a niche in defense technology. News reports refer to partnership agreements between large defense technology corporations such as Northrop Grumman, Lockheed Martin, Raytheon, and startups in Artificial Intelligence/Machine Learning (AI/ML), Image Intelligence (IMINT) Recognition, Battlefield Awareness, Multi‐Domain Battle Management Team (MBMT) Mobility Platform, Tactical Augmented Reality (TAR), Radio Frequency and Wireless Surveillance (RFWS) Systems. At the same time, defense startups like Palantir and Anduril, which have relied on the use of AI since their inception, have now emerged as major game-changing market players. Undoubtedly, this trend will continue in the future and we will soon see significant changes in the direction of enhanced national security and deployed defense technologies. As proof, Project Maven, announced in 2017 according to various sources, is being tested in combat use in military conflicts in Ukraine and Israel. A wearable mobile platform with a high-performance computing module within the Soldier-as-Sensor concept is expected to complement the Battlefield Management System (BMS) and Joint All-Domain Command and Control (JADC2) ecosystems. At least a few defense agencies periodically publish innovation requests for these and related research and technology areas in the prototype stage. Today's highly mobile battlefield threats and higher demands on the cognitive capabilities of operators are driving the need to act simultaneously on multiple fronts. Everyone is required to have accurate Tactical Situational Awareness, Friend-or-Foe Identification, Synthetic Day/Night Vision, BLOS Communications, Sensors, Signal Intelligence, RF Wideband Digital Beamforming, and be an element for sharing this information in peer-to-peer networks. Until recently, some technological limitations have prevented the creation of a convenient mobile platform with dimensions that would allow a person to wear an electronic device for extended periods. For several years, Microsoft tried to adapt its HoloLense 2 device for the IVAS defense project. However, immersive technologies (AR/MR) designed for entertainment, education and training are not well-suited for the military. At Furtherium, my team and I initially created the concept of AI-powered Tactical Situational Awareness Mobility Platform primarily for the military, security forces, rescuers, and firefighters. The key advantages of this technology are its compact design, fast performance, integration of all important functions, and affordable price. And a special feature is the application of several AI technologies on a local device. I founded Furtherium in 2021 and started developing an integrated product ARHUDFM (Augmented Reality Head-Up Display Fullface Mask) - Hardware modules, SaaS and Healthcare modules. Several detection and recognition functions cover visual information in the visible and infrared spectrum, sound and radio frequency waves over a very wide range, passive radar signatures and an EOD scanner for explosive detection. The core of the technology is not only pre-trained Machine Learning models, but also neural network generative AI models that are trained together with each user. For this purpose, advanced CPU and NPU technologies are used in each device. You could say this is the next step in using generative AI in mobile platforms, focused not only on defense applications, with huge dual-use potential. Fine-tuning a local model on a user's device still looks unfamiliar right now, but that's the future. In addition to the key goals of detection and recognition, other basic functions are Omnichannel Messenger, Media Capture and Exchange, Ballistic Fire Assistant and Spotter, Multi-Domain Operations, RC Universal Controller for drones and robots. You don't have to look into the future to understand the trends, just open YouTube and see for yourself the training programs at the U.S. Marine Corps and Army. I'd like to share the new YouTube video we created that demonstrates the capabilities of my startup's mobile platform, controlled primarily by a voice assistant that adapts to each user. https://www.youtube.com/watch?v=kIUt2jbYHMs&embedable=true It seems natural and logical now, but we didn't come to this immediately, it took several years of research into behavioral capabilities, cognitive constraints, tactical documentation and the capabilities of different electronic component architectures. The technical requirements of the military are very different from civilian applications. This applies not only to Hardware, but also to Software, including how generative AI models should be used, what criteria they are required to meet. Many features must be considered when designing for military applications: Night Vision and Augmented Reality without outward illumination, protection against ionizing radiation, main and backup power system, battery life, active cooling, shock wave and fragments exposure, encryption and hopping frequencies for data exchange, reduced traffic while maintaining data quality. It would be wrong to think about the protection of the device and not provide for the protection of the user. For this purpose, several options are created simultaneously for hearing protection and protection against concussion, eye and face protection, respiratory protection, built-in drinking system, automatic tourniquets to stop acute bleeding. Sensors in the user's arms, legs, back, and chest allow health monitoring during exercises, training, and tactical missions. SAM (Serviceable Addressable Market) for this technology estimates exceed $31B and include markets in over 32 NATO countries and its allies. Amid the Research, Development, Test & Evaluation budget of the U.S. DoD alone of $148B this looks promising. Military innovation budgets are on the rise, that's another trend today. My startup's project involves integrating products from companies such as Palantir, Anduril, Teledyne FLIR, Shield AI, Epirus, Elbit Systems, Raytheon, Northrop Grumman, and Thales Group. Combining non-competing systems creates unique synergies. I am also pleased to announce that we are close to raising funds. Investors are expected to include companies with expertise in the Defense and Security area. More Details Furtherium, Inc. | Website | Wiki | Community | Pitch Deck AI technology is firmly carving out a niche in defense technology. News reports refer to partnership agreements between large defense technology corporations such as Northrop Grumman, Lockheed Martin, Raytheon, and startups in Artificial Intelligence/Machine Learning (AI/ML), Image Intelligence (IMINT) Recognition, Battlefield Awareness, Multi‐Domain Battle Management Team (MBMT) Mobility Platform, Tactical Augmented Reality (TAR), Radio Frequency and Wireless Surveillance (RFWS) Systems. At the same time, defense startups like Palantir and Anduril, which have relied on the use of AI since their inception, have now emerged as major game-changing market players. Undoubtedly, this trend will continue in the future and we will soon see significant changes in the direction of enhanced national security and deployed defense technologies. As proof, Project Maven, announced in 2017 according to various sources, is being tested in combat use in military conflicts in Ukraine and Israel. A wearable mobile platform with a high-performance computing module within the Soldier-as-Sensor concept is expected to complement the Battlefield Management System (BMS) and Joint All-Domain Command and Control (JADC2) ecosystems. At least a few defense agencies periodically publish innovation requests for these and related research and technology areas in the prototype stage. Today's highly mobile battlefield threats and higher demands on the cognitive capabilities of operators are driving the need to act simultaneously on multiple fronts. Everyone is required to have accurate Tactical Situational Awareness, Friend-or-Foe Identification, Synthetic Day/Night Vision, BLOS Communications, Sensors, Signal Intelligence, RF Wideband Digital Beamforming, and be an element for sharing this information in peer-to-peer networks. Until recently, some technological limitations have prevented the creation of a convenient mobile platform with dimensions that would allow a person to wear an electronic device for extended periods. For several years, Microsoft tried to adapt its HoloLense 2 device for the IVAS defense project. However, immersive technologies (AR/MR) designed for entertainment, education and training are not well-suited for the military. At Furtherium, my team and I initially created the concept of AI-powered Tactical Situational Awareness Mobility Platform primarily for the military, security forces, rescuers, and firefighters. The key advantages of this technology are its compact design, fast performance, integration of all important functions, and affordable price. And a special feature is the application of several AI technologies on a local device. I founded Furtherium in 2021 and started developing an integrated product ARHUDFM (Augmented Reality Head-Up Display Fullface Mask) - Hardware modules, SaaS and Healthcare modules. Several detection and recognition functions cover visual information in the visible and infrared spectrum, sound and radio frequency waves over a very wide range, passive radar signatures and an EOD scanner for explosive detection. The core of the technology is not only pre-trained Machine Learning models, but also neural network generative AI models that are trained together with each user. For this purpose, advanced CPU and NPU technologies are used in each device. You could say this is the next step in using generative AI in mobile platforms, focused not only on defense applications, with huge dual-use potential. Fine-tuning a local model on a user's device still looks unfamiliar right now, but that's the future. In addition to the key goals of detection and recognition, other basic functions are Omnichannel Messenger, Media Capture and Exchange, Ballistic Fire Assistant and Spotter, Multi-Domain Operations, RC Universal Controller for drones and robots. You don't have to look into the future to understand the trends, just open YouTube and see for yourself the training programs at the U.S. Marine Corps and Army. I'd like to share the new YouTube video we created that demonstrates the capabilities of my startup's mobile platform, controlled primarily by a voice assistant that adapts to each user. the new YouTube video https://www.youtube.com/watch?v=kIUt2jbYHMs&embedable=true https://www.youtube.com/watch?v=kIUt2jbYHMs&embedable=true It seems natural and logical now, but we didn't come to this immediately, it took several years of research into behavioral capabilities, cognitive constraints, tactical documentation and the capabilities of different electronic component architectures. The technical requirements of the military are very different from civilian applications. This applies not only to Hardware, but also to Software, including how generative AI models should be used, what criteria they are required to meet. Many features must be considered when designing for military applications: Night Vision and Augmented Reality without outward illumination, protection against ionizing radiation, main and backup power system, battery life, active cooling, shock wave and fragments exposure, encryption and hopping frequencies for data exchange, reduced traffic while maintaining data quality. It would be wrong to think about the protection of the device and not provide for the protection of the user. For this purpose, several options are created simultaneously for hearing protection and protection against concussion, eye and face protection, respiratory protection, built-in drinking system, automatic tourniquets to stop acute bleeding. Sensors in the user's arms, legs, back, and chest allow health monitoring during exercises, training, and tactical missions. SAM (Serviceable Addressable Market) for this technology estimates exceed $31B and include markets in over 32 NATO countries and its allies. Amid the Research, Development, Test & Evaluation budget of the U.S. DoD alone of $148B this looks promising. Military innovation budgets are on the rise, that's another trend today. My startup's project involves integrating products from companies such as Palantir, Anduril, Teledyne FLIR, Shield AI, Epirus, Elbit Systems, Raytheon, Northrop Grumman, and Thales Group. Combining non-competing systems creates unique synergies. I am also pleased to announce that we are close to raising funds. Investors are expected to include companies with expertise in the Defense and Security area. More Details Furtherium, Inc. | Website | Wiki | Community | Pitch Deck Website Wiki Community Pitch Deck

AI's Influence is Going to Go Deeper Than Just the Corporate Sector

AI এর প্রভাব শুধু কর্পোরেট সেক্টরের চেয়ে গভীরে যেতে চলেছে৷

Der Einfluss der KI wird über den Unternehmenssektor hinausgehen

La influencia de la IA va a llegar más allá del sector corporativo

L’influence de l’IA va s’étendre au-delà du secteur privé

एआई का प्रभाव सिर्फ कॉर्पोरेट क्षेत्र तक ही सीमित नहीं रहेगा

AIの影響は企業部門だけにとどまらない

AI의 영향력은 기업 부문에만 국한되지 않고 더욱 깊어질 것입니다.

A influência da IA vai além do setor corporativo

Влияние ИИ выйдет за рамки корпоративного сектора

Yapay Zekanın Etkisi Sadece Kurumsal Sektörle Sınırlı Kalmayacak

Ảnh hưởng của AI sẽ không chỉ giới hạn trong lĩnh vực doanh nghiệp

人工智能的影响将不仅仅局限于企业领域

Templating in Software Development: Taking a Deeper Look

সফ্টওয়্যার ডেভেলপমেন্টে টেমপ্লেটিং: গভীরভাবে নজর দেওয়া

Templating in der Softwareentwicklung: Ein genauerer Blick

Plantillas en el desarrollo de software: una mirada más profunda

Création de modèles dans le développement de logiciels : examen plus approfondi

सॉफ्टवेयर विकास में टेम्प्लेटिंग: गहराई से देखें

ソフトウェア開発におけるテンプレート化: より深く考える

소프트웨어 개발의 템플릿: 더 자세히 살펴보기

Modelagem no desenvolvimento de software: uma análise mais aprofundada

Шаблоны в разработке программного обеспечения: более глубокий взгляд

Yazılım Geliştirmede Şablon Oluşturma: Daha Derin Bir Bakış

Tạo khuôn mẫu trong phát triển phần mềm: Nhìn sâu hơn

软件开发中的模板：深入了解

Welcome to HackerNoon’s Writing Prompts! Would you like to take a stab at answering some of these questions? Click the log in or create account button on the top right to be redirected back to this template! Cloudflare is a great company that has been innovating and challenging the cloud providers that be. Here is an overview of their products and why you should care. Cloud Security Tools: Cloudflare Turnstile For Bot Prevention Instead of Google Captcha, where to prove you are not a bot, a user would need to select all the buses or cars, Cloudflare has made waves with its free alternative to bot prevention: Cloudflare Turnstile. Released in 2022, but already seeing major adoption, you probably have seen it before. One of the great things about this product is that it does not cause as many users to get frustrated and leave. R2: A Cloud Storage Alternative to S3 AWS, GCP, and Azure all charge egress fees per gigabyte - usually in the range of .05$ to .09 $ per GB. This has led to a lot of caching of the objects, and generally, has slowed innovation surrounding video and VPN products on these clouds. When coupled with Cloudflare CDN, this product can be very powerful. It is ideal for web content delivery, video streaming, static site hosting, and applications that benefit from low egress costs and edge delivery. Cloud Security Solutions: Adaptive DDoS protection One of Cloudflare’s earliest successful products was its DDOS protection. One can imagine that a company that routes 20% of all internet traffic, can have very advanced security tools. The CEO tells a story of how they protected Eurovision from DDOS attacks, and Cloudflare's DDoS protection is often considered superior to other solutions for several reasons. Here are the key factors that contribute to its effectiveness and popularity: Cloudflare’s Web Application Firewall (WAF) Web Application Firewalls that Cloudflare provides have proved to be highly useful for securing public APIs. They can set IP-based throttle limits, and protect against malicious payloads. One of the great things about this cloud security tool is that it is easy to set up and has so much power. From my experience, the first time I noticed it was when I was using a public API for stock history. I was being throttled after 25 calls on my IP address. Cloudflare managed rules


Core OWASP rules (Top 10 exploit patterns)


Advanced rate limiting In the modern world, data is key. There are so many companies (ChatGPT) that scrape the web consistently by using bots. While in the competitive cyberspace, bots will often adapt and find a way, for the majority of naive scraping tools, Cloudflare is able to block them. Conclusion At the end of the day, we all assume that the major clouds have ‘figured it out’. From my software development experience in Big Tech, there is a lot of misguided designs and principles. All of us have grown up on the internet, have seen it change human dialogue, the way we think about the world, and has entertained us for years. The most shocking thing about Cloudflare is that they don’t discriminate or judge their customers and they don’t overcharge their customers. If you would like to follow me for more, here is my Cloud security substack, cheers! Welcome to HackerNoon’s Writing Prompts! Would you like to take a stab at answering some of these questions? Click the log in or create account button on the top right to be redirected back to this template! Welcome to HackerNoon ’s Writing Prompts! Would you like to take a stab at answering some of these questions? Click the log in or create account button on the top right to be redirected back to this template! HackerNoon Cloudflare is a great company that has been innovating and challenging the cloud providers that be. Here is an overview of their products and why you should care. Cloud Security Tools: Cloudflare Turnstile For Bot Prevention Instead of Google Captcha, where to prove you are not a bot, a user would need to select all the buses or cars, Cloudflare has made waves with its free alternative to bot prevention: Cloudflare Turnstile . Released in 2022, but already seeing major adoption, you probably have seen it before. One of the great things about this product is that it does not cause as many users to get frustrated and leave. Cloudflare Turnstile Cloudflare Turnstile R2: A Cloud Storage Alternative to S3 AWS, GCP, and Azure all charge egress fees per gigabyte - usually in the range of .05$ to .09 $ per GB. This has led to a lot of caching of the objects, and generally, has slowed innovation surrounding video and VPN products on these clouds. When coupled with Cloudflare CDN, this product can be very powerful. It is ideal for web content delivery, video streaming, static site hosting, and applications that benefit from low egress costs and edge delivery. Cloud Security Solutions: Adaptive DDoS protection One of Cloudflare’s earliest successful products was its DDOS protection. One can imagine that a company that routes 20% of all internet traffic, can have very advanced security tools. The CEO tells a story of how they protected Eurovision from DDOS attacks, and Cloudflare's DDoS protection is often considered superior to other solutions for several reasons. Here are the key factors that contribute to its effectiveness and popularity: Cloudflare’s Web Application Firewall (WAF) Web Application Firewalls that Cloudflare provides have proved to be highly useful for securing public APIs. They can set IP-based throttle limits, and protect against malicious payloads. One of the great things about this cloud security tool is that it is easy to set up and has so much power. From my experience, the first time I noticed it was when I was using a public API for stock history. I was being throttled after 25 calls on my IP address. Web Application Firewalls Web Application Firewalls cloud security tool cloud security tool Cloudflare managed rules Core OWASP rules (Top 10 exploit patterns) Advanced rate limiting Cloudflare managed rules Cloudflare managed rules Core OWASP rules (Top 10 exploit patterns) Core OWASP rules (Top 10 exploit patterns) Advanced rate limiting Advanced rate limiting In the modern world, data is key. There are so many companies (ChatGPT) that scrape the web consistently by using bots. While in the competitive cyberspace, bots will often adapt and find a way, for the majority of naive scraping tools, Cloudflare is able to block them. Conclusion At the end of the day, we all assume that the major clouds have ‘figured it out’. From my software development experience in Big Tech, there is a lot of misguided designs and principles. All of us have grown up on the internet, have seen it change human dialogue, the way we think about the world, and has entertained us for years. The most shocking thing about Cloudflare is that they don’t discriminate or judge their customers and they don’t overcharge their customers. If you would like to follow me for more, here is my Cloud security substack, cheers! Cloud security

A Deeper Look at the Curious Case of Cloudflare

AI is one of the hottest and most controversial topics under discussion in the tech community. Most believe that it has progressed too quickly and without any real input from the public. Stack Overflow users are furious that their responses are going to be used to train ChatGPT without consent, via an API agreement. Elon Musk, who had a hand in building ChatGPT, also believes that it is one of the biggest existential threats to humanity. Many business executives, on the other hand, believe that it’s just a paradigm shift and will ultimately create more roles, freeing humans to be more creative. At no point in history has technological progress ever resulted in fewer jobs, but best of luck explaining that to the masses. Too many people, apparently, are taking their cues from movies where Skynet will become alive and take the planet over. The truth lies somewhere in the middle, and there are much deeper perspectives to be explored. AI is primarily an existential issue tied to our daily lifestyles. The Real Role Of AI In A Misaligned Public Order The problem is not AI in and of itself. The problem is actually how human beings approach their lives in general. There is a global attitude that the purpose of life is to make as much money as possible and be a productive working citizen. So, it was only a matter of time before humans created a machine that was thousands of times more productive than they were, leading them to question the nature of existence and their purpose for being. The belief of “AI taking all the jobs” is a very childish one that has, unfortunately, gained sway in the media and the public consciousness. People, for the most part, work jobs they hate, and are full of fear when AI threatens to take them away. It’s a survival mode and a mindset that needs to be replaced. The purpose of life is not productivity, but joy and creativity. When people understand that, their perspective on life can completely change. AI will take all the boring and monotonous jobs, and free us up for better work which enhances the meaning of life - joy and creativity. What AI Can’t Do There are many things that AI can’t do accurately, at least right now. Most of these are roles that involve emotional awareness, social interaction, and situational adaptation. Another purpose of human existence is emotional connection, which provides life’s greatest blessings and challenges. AI has no capacity in this regard and in many industry sectors. AI lacks dexterity for jobs like plumbing, carpentry, electrical work, and many other aspects of construction. But it also lacks the social skills to interact with people in their homes when coming to complete a job. In other words, you simply can’t slot AI into existing jobs without factoring in the human elements into them. Most of the research reports/studies are entirely analytical - as if jobs are completed without any creative elements or social interaction. Even areas where AI should excel might be suspect. For instance, HFT bots cannot spot truly unique trading opportunities and are only as useful as inputs. Price forecasting analyst Taki Tsaklanos believes that AI is unable to predict an XRP breakout: “I know how to analyze the XRP chart in a way AI cannot. It is simple for me: AI is absolutely clueless when it comes to predicting XRP” The idea of set-and-forget AI is ridiculously naive. AI is merely an accelerator for processes. Processes and inputs need to be refined so that the AI can succeed. If you give AI bad parameters, it will speed up the rate of failure. We need quality content, not more content. It’s about producing higher quality, not more meaningless stuff. Problems That Actually Exist Of course, there still remains a question of how AI is to be implemented, and the pace of introduction needs to slow down. The power of AI to create realistic deepfakes is already extremely troubling. There are many real issues with AI - “taking all the jobs” is at the absolute bottom of the list. There is also an alienation issue. People are becoming more and more disconnected from each other. You can’t beat face-to-face interaction and real, meaningful, friendships and experiences, as well as a strong community of live people. AI needs to be used as a tool to enhance life, instead of bypassing it with mindless automation. While it’s great from the perspective of productivity, we’re not actually in a productivity crisis and never have been. AI is one of the hottest and most controversial topics under discussion in the tech community. Most believe that it has progressed too quickly and without any real input from the public. Stack Overflow users are furious that their responses are going to be used to train ChatGPT without consent, via an API agreement. used to train ChatGPT used to train ChatGPT Elon Musk, who had a hand in building ChatGPT, also believes that it is one of the biggest existential threats to humanity. Many business executives, on the other hand, believe that it’s just a paradigm shift and will ultimately create more roles, freeing humans to be more creative. At no point in history has technological progress ever resulted in fewer jobs, but best of luck explaining that to the masses. Too many people, apparently, are taking their cues from movies where Skynet will become alive and take the planet over. At no point in histor y At no point in histor The truth lies somewhere in the middle, and there are much deeper perspectives to be explored. AI is primarily an existential issue tied to our daily lifestyles. The Real Role Of AI In A Misaligned Public Order The problem is not AI in and of itself. The problem is actually how human beings approach their lives in general. There is a global attitude that the purpose of life is to make as much money as possible and be a productive working citizen. So, it was only a matter of time before humans created a machine that was thousands of times more productive than they were, leading them to question the nature of existence and their purpose for being. The belief of “AI taking all the jobs” is a very childish one that has, unfortunately, gained sway in the media and the public consciousness. People, for the most part, work jobs they hate, and are full of fear when AI threatens to take them away. It’s a survival mode and a mindset that needs to be replaced. “AI taking all the jobs” The purpose of life is not productivity, but joy and creativity. When people understand that, their perspective on life can completely change. AI will take all the boring and monotonous jobs, and free us up for better work which enhances the meaning of life - joy and creativity. What AI Can’t Do There are many things that AI can’t do accurately, at least right now. Most of these are roles that involve emotional awareness, social interaction, and situational adaptation. Another purpose of human existence is emotional connection, which provides life’s greatest blessings and challenges. AI has no capacity in this regard and in many industry sectors. many industry sectors. AI lacks dexterity for jobs like plumbing, carpentry, electrical work, and many other aspects of construction. But it also lacks the social skills to interact with people in their homes when coming to complete a job. In other words, you simply can’t slot AI into existing jobs without factoring in the human elements into them. Most of the research reports/studies are entirely analytical - as if jobs are completed without any creative elements or social interaction. Even areas where AI should excel might be suspect. For instance, HFT bots cannot spot truly unique trading opportunities and are only as useful as inputs. Price forecasting analyst Taki Tsaklanos believes that AI is unable to predict an XRP breakout : AI is unable to predict an XRP breakout AI is unable to predict an XRP breakout “I know how to analyze the XRP chart in a way AI cannot. It is simple for me: AI is absolutely clueless when it comes to predicting XRP” “I know how to analyze the XRP chart in a way AI cannot. It is simple for me: AI is absolutely clueless when it comes to predicting XRP” “I know how to analyze the XRP chart in a way AI cannot. It is simple for me: AI is absolutely clueless when it comes to predicting XRP” The idea of set-and-forget AI is ridiculously naive. AI is merely an accelerator for processes. Processes and inputs need to be refined so that the AI can succeed. If you give AI bad parameters, it will speed up the rate of failure. We need quality content, not more content. It’s about producing higher quality, not more meaningless stuff. Problems That Actually Exist Of course, there still remains a question of how AI is to be implemented, and the pace of introduction needs to slow down. The power of AI to create realistic deepfakes is already extremely troubling. There are many real issues with AI - “taking all the jobs” is at the absolute bottom of the list. AI to create realistic deepfakes “taking all the jobs” There is also an alienation issue. People are becoming more and more disconnected from each other. You can’t beat face-to-face interaction and real, meaningful, friendships and experiences, as well as a strong community of live people. AI needs to be used as a tool to enhance life, instead of bypassing it with mindless automation. While it’s great from the perspective of productivity, we’re not actually in a productivity crisis and never have been.

A Deeper Perspective On AI - Why A Paradigm Shift Is Needed

This paper is available on arxiv under CC 4.0 license.
Authors:
(1) Kiriaki Frangias;
(2) Andrew Lin;
(3) Ellen Vitercik;
(4) Manolis Zampetakis. Table of Links Abstract and Introduction Warm-up: Agents with known equal disutility Agents with unknown disutilites Experiments Conclusions and future directions, References A Summary of notation B Omitted proofs from Section 2 C Omitted proofs from Section 3 D Additional Information about Experiments 4 Experiments In the following section, we present experiments that test our models and algorithm. We evaluate the impact of our algorithm on the principal’s utility compared to when she chooses not to incentivize agents, the efficacy of our algorithm in returning correct pairwise comparisons, and the robustness of our model to noisy values of ψ and π. Experimental procedure. Unless otherwise specified, we keep the following set of hyperparameter values across our experiments: π = 0.8, δ = 0.01, s = 100, n = 100, ψ¯ = 2, and ψ = 0.01. Peer grading is an example of a setting where s = n since the number of assignments is the same as the number of students. In Appendix D, we include experiments with a broad variety of parameter settings. Since we add noise to our model in these experiments, CrowdSort may not catch all bad agents and thus may return conflicting comparisons. We discard all conflicting comparisons (see Algorithm 4 in Appendix D). Despite this pessimistic post-processing strategy, our experiments show that we can recover most comparisons even under noise. 4.1 Principal’s utility The empirical utilities are calculated as follows. Let m be the number of agents not identified as “bad” by CrowdSort and k be the number of comparisons after post-processing. We set λ = 2 to be the principal’s per-comparison utility and define to be her empirical utility for running CrowdSort with p∗, v∗ , and r∗ defined in Equations (1) and (2). Similarly is her expected utility for performing all comparisons herself, where 2n log n is the expected number of comparisons required by quicksort (this value could easily be adjusted for any sorting algorithm). For ease of comparison, the y-axis is scaled by dividing all values by the baseline utility U0. We note that the size of the confidence interval decreases with π. For small values of π, a small number of agents are “good”, even though all agents exert effort. Then with higher probability, k will be small. This corresponds to higher variance in the principal’s utility across trials. In Figure 1b, we add noise to the parameter ψ and analyze the effect on the principal’s utility. We sample the disutility of each agent independently and identically from the distribution Uniform(ψ − 0.5ψ, ψ + 0.5ψ). However, we still calculate p∗, v∗, and r∗ from Equations (1) and (2) using ψ itself instead of the noisy version. The blue shading corresponds to a 90% confidence interval across 50 trials. Discussion. Comparing Figures 1a and 1b, we can see that even when noise is added to our model, our theoretical threshold is close to the point at which the principal’s utility from running CrowdSort dips below their utility from performing all comparisons herself. This demonstrates that our theoretical results are robust to noise in the model. When adding noise to ψ, the principal must pay fewer agents because fewer are incentivized to exert effort. Thus, we see a steeper decline in Figure 1a compared to 1b. 4.2 Number of pairwise comparisons returned Discussion. These experiments show that our algorithm retrieves all comparisons in the absence of noise. Moreover, even when a large amount of noise is added to ψ or π, our algorithm returns almost all of the comparisons with sufficient payment, despite the postprocessing phase excluding any conflicting comparison, again illustrating our algorithm’s robustness. 4.3 Optimization problem Discussion. For suitable values of g, approximately 40 samples are required for the utility of the principal under our model to surpass the utility she would get, if she were to perform all pairwise comparisons herself. We also notice that the principal’s utility increases as we increase either g or the number of samples. This result is shown in the multi-colored plots in Figure 3. Therefore, even for a small number of samples, the principal’s utility when implementing CrowdSort is well above her utility if she performed all comparisons on her own. This paper is available on Arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. Authors: (1) Kiriaki Frangias; (2) Andrew Lin; (3) Ellen Vitercik; (4) Manolis Zampetakis. This paper is available on arxiv under CC 4.0 license. Authors: Authors: (1) Kiriaki Frangias; (2) Andrew Lin; (3) Ellen Vitercik; (4) Manolis Zampetakis. Table of Links Abstract and Introduction Abstract and Introduction Warm-up: Agents with known equal disutility Warm-up: Agents with known equal disutility Agents with unknown disutilites Agents with unknown disutilites Experiments Experiments Conclusions and future directions, References Conclusions and future directions, References A Summary of notation A Summary of notation B Omitted proofs from Section 2 B Omitted proofs from Section 2 C Omitted proofs from Section 3 C Omitted proofs from Section 3 D Additional Information about Experiments D Additional Information about Experiments 4 Experiments In the following section, we present experiments that test our models and algorithm. We evaluate the impact of our algorithm on the principal’s utility compared to when she chooses not to incentivize agents, the efficacy of our algorithm in returning correct pairwise comparisons, and the robustness of our model to noisy values of ψ and π. Experimental procedure. Unless otherwise specified, we keep the following set of hyperparameter values across our experiments: π = 0.8, δ = 0.01, s = 100, n = 100, ψ¯ = 2, and ψ = 0.01. Peer grading is an example of a setting where s = n since the number of assignments is the same as the number of students. In Appendix D, we include experiments with a broad variety of parameter settings. Experimental procedure. Since we add noise to our model in these experiments, CrowdSort may not catch all bad agents and thus may return conflicting comparisons. We discard all conflicting comparisons (see Algorithm 4 in Appendix D). Despite this pessimistic post-processing strategy, our experiments show that we can recover most comparisons even under noise. 4.1 Principal’s utility The empirical utilities are calculated as follows. Let m be the number of agents not identified as “bad” by CrowdSort and k be the number of comparisons after post-processing. We set λ = 2 to be the principal’s per-comparison utility and define to be her empirical utility for running CrowdSort with p∗, v∗ , and r∗ defined in Equations (1) and (2). Similarly is her expected utility for performing all comparisons herself, where 2n log n is the expected number of comparisons required by quicksort (this value could easily be adjusted for any sorting algorithm). For ease of comparison, the y-axis is scaled by dividing all values by the baseline utility U0. We note that the size of the confidence interval decreases with π. For small values of π, a small number of agents are “good”, even though all agents exert effort. Then with higher probability, k will be small. This corresponds to higher variance in the principal’s utility across trials. In Figure 1b, we add noise to the parameter ψ and analyze the effect on the principal’s utility. We sample the disutility of each agent independently and identically from the distribution Uniform(ψ − 0.5ψ, ψ + 0.5ψ). However, we still calculate p∗, v∗, and r∗ from Equations (1) and (2) using ψ itself instead of the noisy version. The blue shading corresponds to a 90% confidence interval across 50 trials. Discussion . Comparing Figures 1a and 1b, we can see that even when noise is added to our model, our theoretical threshold is close to the point at which the principal’s utility from running CrowdSort dips below their utility from performing all comparisons herself. This demonstrates that our theoretical results are robust to noise in the model. Discussion When adding noise to ψ, the principal must pay fewer agents because fewer are incentivized to exert effort. Thus, we see a steeper decline in Figure 1a compared to 1b. 4.2 Number of pairwise comparisons returned Discussion . These experiments show that our algorithm retrieves all comparisons in the absence of noise. Moreover, even when a large amount of noise is added to ψ or π, our algorithm returns almost all of the comparisons with sufficient payment, despite the postprocessing phase excluding any conflicting comparison, again illustrating our algorithm’s robustness. Discussion 4.3 Optimization problem Discussion . For suitable values of g, approximately 40 samples are required for the utility of the principal under our model to surpass the utility she would get, if she were to perform all pairwise comparisons herself. We also notice that the principal’s utility increases as we increase either g or the number of samples. This result is shown in the multi-colored plots in Figure 3. Therefore, even for a small number of samples, the principal’s utility when implementing CrowdSort is well above her utility if she performed all comparisons on her own. Discussion This paper is available on Arxiv under CC 4.0 license. This paper is available on Arxiv under CC 4.0 license. Arxiv

Algorithmic Contract Design for Crowdsourced Ranking: A Deeper Look at Our Experiments

This paper is available on arxiv under CC 4.0 license.
Authors:
(1) Falke B. Ø. Carlsen, Department of Computer Science, Aalborg University, Denmark & falkeboc@cs.aau.dk;
(2) Lars Bo P. Frydenskov, Department of Computer Science, Aalborg University, Denmark & larsbopark@gmail.com;
(3) Nicolaj Ø. Jensen, Department of Computer Science, Aalborg University, Denmark & noje@cs.aau.dk;
(4) Jener Rasmussen, jener@jener.dk;
(5) Mathias M. Sørensen, mathiasmehlsoerensen@gmail.com;
(6) Asger G. Weirsøe, asger@weircon.dk;
(7) Mathias C. Jensen, Department of Computer Science, Aalborg University, Denmark & mcje@cs.aau.dk;
(8) Kim G. Larsen, Department of Computer Science, Aalborg University, Denmark & kgl@cs.aau.dk. Table of Links Introduction Definitions Model Checking Tool Overview Evaluation Conclusion and References 3 Model Checking In order to check if a CGS satisfies an ATL property, CGAAL encodes the problem as an extended dependency graph and finds a fixed-point assignment describing the satisfaction relation. Extended Dependency Graphs An extended dependency graph (EDG) is a tuple G = ⟨C,E,N⟩ where C is a finite set of configurations (vertices), E ⊆C×P(C) is a finite set of hyper-edges, and N ⊆C×C is a finite set of negation edges. 3.1 Encoding of ATL in an EDG We first establish some definitions related to subsets of move vectors induced by a coalition of players. With this notation in mind, we define the function pmoves. The function pmoves gives the set of all possible partial moves that follow from the players of the set A ⊆ Σ making a combination of moves in the state q ∈ Q: For instance, if D(q) = {1,2}×{1,2} in a 2-player game and A = {1}. Then pmoves(q,A) = {{1}× {1,2},{2}×{1,2}} contains two partial moves. One where player 1 chooses action 1, and another where player 1 chooses action 2. In other words, the pmoves function constructs the partial moves that the set A of players can choose from when working together in the state q. If A chooses V ∈ pmoves(q,A), then V contains all move vectors that result from the remaining players Σ/A also making a choice. We define a partial transition function ∆ that produces a set of possible successor states given a state q ∈ Q and a partial move V. That is With EDGs, assignments, and partial moves introduced, we can now define how the satisfaction relation ⊧ can be encoded as an EDG. Encoding. Given a CGS S, a state q of S, and an ATL state formula φ we now construct an EDG where every configuration is either a pair of a state and a formula, or a triple of a state, partial move, and a formula. The triples represent partially evaluated states, where some players’ moves are already set. Starting from the initial pair ⟨q,φ⟩, the dependency graph is constructed according to Figures 1 to 6. The figures show which outgoing edges each configuration has and the target configurations of those edges. The proof of Theorem 3.1 can be found in the extended version of this paper. 3.2 Search Strategies The CERTAINZERO algorithm is controlled by a search strategy, which determines in which order the edges are explored and evaluated. Our search strategies for CGAAL include the common breadth-first search (BFS) and depth-first search (DFS) strategies, as well as multiple search strategies based on heuristics. Some of these are discussed in the following subsections. Which strategy is best depends heavily on the shape of the EDG which is determined by the CGS and the ATL formula in question. An evaluation of the strategies can be found in Section 5. The BFS strategy is the default strategy. Dependency Heuristic Search (DHS) PageRank [15] is an algorithm that was created to estimate the importance of a website based on how many other websites have links to it. The idea has since been used in other areas, such as graph recommendation systems [13] or measuring structural context similarity with SimRank [7]. Our dependency heuristic search (DHS) uses a similar idea by assuming that configurations with many ingoing edges are important. Finding the assignments of these, results in more back-propagation, bringing us closer to termination. In other words, the heuristic focuses on the trunk of the EDG where certain assignments are of high value. Specifically, DHS prioritise edges where the source configuration has a high number of ingoing edges. That is, if e is an edge with source configuration c then: Edges with the same priority are subject to FIFO ordering. However, since we do not know the whole graph in advance, we must continuously update the priority of edges when we explore successors of new configurations. This is only a small overhead with a priority queue data structure. We also support an alternative search strategy involving linear programming. It is called Linear Representative Search (LRS) and it computes the distance described above for the root configuration only. The closest satisfying state s is then saved and edges are prioritised based on the 1-norm distance between s and the state in their source configuration. In other words, we assume that the state s found for the root represents all satisfied states. As a result this search strategy is cheaper than LPS but risks being inaccurate. Instability Heuristic Search (IHS) The Petri Net model checker TAPAAL [8] implements the CERTAINZERO algorithm as well. Their configurations consist of a marking (a state) and a property, and their default search strategy uses a heuristic that estimates the distance between the marking of the configuration and a marking that satisfies the formula of the configuration. Our novel instability heuristic is inspired by their heuristic, but we differ by acknowledging that since there are negation edges in the EDG, we may not always be looking for a state that satisfies the formula. That is, if the state already satisfies the associated formula, we estimate the distance to a state that does not instead. This guides the search towards configurations where the assignment is unstable and thus may have a high influence. Algorithms 1 and 2 find the distIHS described above for an edge. The algorithms use an abstract metric BiDist which for state-property pairs finds ⟨tˆ, ˆf⟩ where tˆ is the distance to the property being true and ˆf is the distance to the property being false. This paper is available on Arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. Authors: (1) Falke B. Ø. Carlsen, Department of Computer Science, Aalborg University, Denmark & falkeboc@cs.aau.dk; (2) Lars Bo P. Frydenskov, Department of Computer Science, Aalborg University, Denmark & larsbopark@gmail.com; (3) Nicolaj Ø. Jensen, Department of Computer Science, Aalborg University, Denmark & noje@cs.aau.dk; (4) Jener Rasmussen, jener@jener.dk; (5) Mathias M. Sørensen, mathiasmehlsoerensen@gmail.com; (6) Asger G. Weirsøe, asger@weircon.dk; (7) Mathias C. Jensen, Department of Computer Science, Aalborg University, Denmark & mcje@cs.aau.dk; (8) Kim G. Larsen, Department of Computer Science, Aalborg University, Denmark & kgl@cs.aau.dk. This paper is available on arxiv under CC 4.0 license. Authors: Authors: (1) Falke B. Ø. Carlsen, Department of Computer Science, Aalborg University, Denmark & falkeboc@cs.aau.dk; (2) Lars Bo P. Frydenskov, Department of Computer Science, Aalborg University, Denmark & larsbopark@gmail.com; (3) Nicolaj Ø. Jensen, Department of Computer Science, Aalborg University, Denmark & noje@cs.aau.dk; (4) Jener Rasmussen, jener@jener.dk; (5) Mathias M. Sørensen, mathiasmehlsoerensen@gmail.com; (6) Asger G. Weirsøe, asger@weircon.dk; (7) Mathias C. Jensen, Department of Computer Science, Aalborg University, Denmark & mcje@cs.aau.dk; (8) Kim G. Larsen, Department of Computer Science, Aalborg University, Denmark & kgl@cs.aau.dk. Table of Links Introduction Introduction Definitions Definitions Model Checking Model Checking Tool Overview Tool Overview Evaluation Evaluation Conclusion and References Conclusion and References 3 Model Checking In order to check if a CGS satisfies an ATL property, CGAAL encodes the problem as an extended dependency graph and finds a fixed-point assignment describing the satisfaction relation. Extended Dependency Graphs An extended dependency graph (EDG) is a tuple G = ⟨C,E,N⟩ where Extended Dependency Graphs extended dependency graph G = ⟨C,E,N⟩ C is a finite set of configurations (vertices), E ⊆C×P(C) is a finite set of hyper-edges, and N ⊆C×C is a finite set of negation edges. C E ⊆C×P(C) N ⊆C×C 3.1 Encoding of ATL in an EDG We first establish some definitions related to subsets of move vectors induced by a coalition of players. With this notation in mind, we define the function pmoves . The function pmoves gives the set of all pmoves pmoves possible partial moves that follow from the players of the set A ⊆ Σ making a combination of moves in the state q ∈ Q : A ⊆ Σ q ∈ Q For instance, if D(q) = {1,2}×{1,2} in a 2-player game and A = {1}. Then pmoves (q,A) = {{1}× {1,2},{2}×{1,2}} contains two partial moves. One where player 1 chooses action 1, and another where player 1 chooses action 2. In other words, the pmoves function constructs the partial moves that the set A of players can choose from when working together in the state q. If A chooses V ∈ pmoves (q,A), then V contains all move vectors that result from the remaining players Σ/A also making a choice. pmoves pmoves pmoves We define a partial transition function ∆ that produces a set of possible successor states given a state q ∈ Q and a partial move V . That is partial transition function q ∈ Q V With EDGs, assignments, and partial moves introduced, we can now define how the satisfaction relation ⊧ can be encoded as an EDG. Encoding . Given a CGS S, a state q of S, and an ATL state formula φ we now construct an EDG where every configuration is either a pair of a state and a formula, or a triple of a state, partial move, and a formula. The triples represent partially evaluated states, where some players’ moves are already set. Starting from the initial pair ⟨q,φ⟩, the dependency graph is constructed according to Figures 1 to 6. The figures show which outgoing edges each configuration has and the target configurations of those edges. Encoding The proof of Theorem 3.1 can be found in the extended version of this paper. 3.2 Search Strategies The CERTAINZERO algorithm is controlled by a search strategy, which determines in which order the edges are explored and evaluated. Our search strategies for CGAAL include the common breadth-first search (BFS) and depth-first search (DFS) strategies, as well as multiple search strategies based on heuristics. Some of these are discussed in the following subsections. Which strategy is best depends heavily on the shape of the EDG which is determined by the CGS and the ATL formula in question. An evaluation of the strategies can be found in Section 5. The BFS strategy is the default strategy. Dependency Heuristic Search (DHS) PageRank [15] is an algorithm that was created to estimate the importance of a website based on how many other websites have links to it. The idea has since been used in other areas, such as graph recommendation systems [13] or measuring structural context similarity with SimRank [7]. Our dependency heuristic search (DHS) uses a similar idea by assuming that configurations with many ingoing edges are important. Finding the assignments of these, results in more back-propagation, bringing us closer to termination. In other words, the heuristic focuses on the trunk of the EDG where certain assignments are of high value. Specifically, DHS prioritise edges where the source configuration has a high number of ingoing edges. That is, if e is an edge with source configuration c then: Dependency Heuristic Search (DHS) trunk e c Edges with the same priority are subject to FIFO ordering. However, since we do not know the whole graph in advance, we must continuously update the priority of edges when we explore successors of new configurations. This is only a small overhead with a priority queue data structure. We also support an alternative search strategy involving linear programming. It is called Linear Representative Search (LRS) and it computes the distance described above for the root configuration only. The closest satisfying state s is then saved and edges are prioritised based on the 1-norm distance between s and the state in their source configuration. In other words, we assume that the state s found for the root represents all satisfied states. As a result this search strategy is cheaper than LPS but risks being inaccurate. Instability Heuristic Search (IHS) The Petri Net model checker TAPAAL [8] implements the CERTAINZERO algorithm as well. Their configurations consist of a marking (a state) and a property, and their default search strategy uses a heuristic that estimates the distance between the marking of the configuration and a marking that satisfies the formula of the configuration. Our novel instability heuristic is inspired by their heuristic, but we differ by acknowledging that since there are negation edges in the EDG, we may not always be looking for a state that satisfies the formula. That is, if the state already satisfies the associated formula, we estimate the distance to a state that does not instead. This guides the search towards configurations where the assignment is unstable and thus may have a high influence. Algorithms 1 and 2 find the distIHS described above for an edge. The algorithms use an abstract metric BiDist which for state-property pairs finds ⟨tˆ, ˆf⟩ where tˆ is the distance to the property being true and ˆf is the distance to the property being false. Instability Heuristic Search (IHS) This paper is available on Arxiv under CC 4.0 license. This paper is available on Arxiv under CC 4.0 license. Arxiv

How CGAAL Model-Checks: A Deeper Insight

Object-Centric Models for ARC Grids: A Deeper Look

This article is a continuation of my discussion on the tokenization of real assets. In this piece, I delve into the technical intricacies of the ERC-3643 standard, highlighting its distinctions from the ERC-20 standard and presenting various use cases. If you missed the first part, you can find it here. https://hackernoon.com/tokenizing-real-world-assets-rwa-part-1-a-deep-dive-into-erc-3643?embedable=true Overview ERC-3643 One of the key issues that need to be resolved before RWAs can be safely tokenized is the legal compliance of their transfer and transactions. In other words, they need to be regulated. The ERC-3643 standard is a tokenization standard for RWAs that simplifies their transfer to meet regulatory requirements. It is designed to make it possible to tokenize RWAs at the institutional level. For better clarity, we will compare the ERC-3643 standard to the ERC-20 standard because the ERC-3643 standard is actually another version of the ERC-20. is a fungible token standard that requires implementation of all the functions that the standard is designed to implement, including token transfer functions and functions that return address balances. ERC-20 In the case of ERC-20, the standard does not require token transfer restrictions. This means that in most cases, anyone can own these tokens, even if it is not known exactly who owns them, so transfer and transact with them without restrictions. These tokens are called permissionless. Examples of such tokens are , DAI, etc. USDT In the case of the implementation of the ERC-3643 standard, it implies being permissioned. This means that tokens can only be transferred if all conditions for the transfer are met (e.g. participants are verified and other requirements are met), otherwise, it is not possible. The implementation of the ERC-3643 standard also requires a function that is responsible for the recovery of tokens, for example, in case of loss of access to the account. The ERC-20 standard does not require the implementation of such a function, which means that in case of loss of keys to the account access to tokens can be lost without the possibility of recovery. In the case of ERC-3643, if access to the account is lost, the tokens can be restored to another account due to the required implementation of the recovery function. The ERC-3643 standard means that accounts that will be able to transact in tokens must be identified, as required by the laws of most countries.  With the ERC-20 standard, we have no such requirement, which means that all accounts making transactions remain impersonal due to the lack of identification and such transactions cannot be verified. In the case of ERC-3643, by requiring all accounts to be identified, the blockchain can now become a verifiable ledger (by knowing who is behind each account), and storing data on it will only increase security compared to the current infrastructure. The ERC-3643 standard is designed for Ethereum networks and allows for the creation of programmable agreements for issuing, managing, and trading RWA tokens. All the data required for this is stored directly in the blockchain, such as the addresses of the owners, conditions for token transfer, account status, etc. Because smart contracts calculate and process data, it is all done automatically, with no need for manual verification. For example, let's compare a transaction using an ERC-20 token and a transaction with ERC-3643. In the case of ERC-20, one party calls a function in the token smart contract and there is a transfer of the token to the other party that was specified in that transaction. In the case of ERC-3643, a party makes a token transfer function call to a smart contract, but in this case, there is no immediate token transfer. First, there is a check of all the conditions that are set for such a token transfer, usually on another contract, which processes all the necessary data and returns a status that reflects whether the token transfer can be completed or not. If this status indicates that all the necessary conditions are met, only in this case the token transfer from one party to another is executed. Essential Contracts of ERC-3643 Implementation Now let's look at the essential contracts of the ERC-3643 implementation and what each is responsible for: Each user is assigned a contract that is responsible for processing their identity and storing all claims associated with it. Each user is represented by an Identity contract. Identity: This is where the various claim topics that can be added or deleted to the contract Identity are managed. The owner is responsible for managing these registries per his claims. Claim Topics Registry: Represents a registry of contract addresses that are trusted issuers of claims. This is to ensure that claims are only issued by these trusted issuers. Trusted Issuers Registry: Performs a check on Identities against the rule module it contains. This check is needed to ensure that only participants who meet the specified requirements can make deals or token transfers. Compliance: Keeps track of all Identities in the system and stores them all in one place. This provides convenient verification and validation of user identity. Identity Registry: How does token transfer control work in ERC-3643? To better illustrate the key point of how token transfers between addresses are controlled to meet the requirements, we will look at some of the key features of the Token For Regulated EXchanges (T-REX) protocol token contract, which is an implementation of the ERC-3643 standard. It is ERC-20 compliant and therefore contains all the functionalities required by the ERC-20 standard. Before we move on to the token transfer function itself, let's look at the other aspects that we need to understand before we do so. There is a special role - Agent, this role is given to an authorized account that performs operational activities. An account with the agent role can freeze addresses, freeze tokens, pause token transfers, issue, burn tokens, etc. We will look at a few of these functions that directly relate to the token transfer function itself. - This function in the token contract is responsible for setting the status for the address (of the user) - whether it is frozen or not. It can be called only by an account with the role. So, if passed - it will mean that this address will be considered frozen, and if - it will be considered unfrozen. setAddressFrozen Agent true false these two functions are responsible for freezing or unfreezing a specific amount of tokens at a specific address ( of the user ). After that the frozen amount of tokens can no longer be transferred, or the number of frozen tokens will be reduced through unfreezing. Available for invocation only for accounts with the role. freezePartialTokens / unfreezePartialTokens - Agent - accepts the and the number of tokens to be frozen. Then it is checked that the total amount of frozen tokens after freezing will not exceed the current token balance of this address. If this condition is reached, the amount of frozen tokens at this address is increased by the . freezePartialTokens user address amount - here we first check that the amount of unfreeze tokens does not exceed the amount of frozen tokens at this address. If the condition is met, the amount of frozen tokens is reduced by the . unfreezePartialTokens amount - responsible for the opportunity to pause or unpause transfers. This functionality is available only for accounts with the role. When calling these functions, the value for state variable is set, if - transfers are paused, if - transfers are not paused. Pause/unpause Agent _tokenPaused true false - this function is responsible for transferring tokens from one address to another. It accepts only two arguments: - the address of the recipient, - the amount to be transferred. This function has no restrictions as to who can call it. But its call is available only if transfers are not paused by the role, for this the modifier is responsible. Transfer to amount Agent whenNotPaused First, it is checked that the recipient address and the sender address (sender is the transaction sender) are not frozen. As we have already seen, is responsible for freezing accounts, and only an account with the role can do it. If at least one of the transfer participants has a frozen account, the transaction will revert with an error. setAddressFrozen Agent Then it is checked that the current available balance of the sender, which is calculated as the difference between the current balance and the amount of frozen tokens, does not exceed the . amount The next check consists of two conditions at once: The function is called in the contract, which is the registry for Identity. checks that the address of the Identity contract (for the recipient account) is listed in the , checks the validity of the that are listed in Identity and returns state based on these checks. Verification of claimTopics includes requesting them from the contract and checking that these are issued by the registry of which is stored in the . isVerified IndetityRegistry isVerified IdentityRegistry claimTopics ClaimTopicsRegistry ClaimTopics TrustedIssuers ClaimTopicsRegistry The next condition to be done is to verify that the sender, recipient and the amount of tokens transferred are in compliance with the offer rules, which are placed in the contract and can be added as modules or created completely custom from scratch. It can define conditions such as the maximum number of investors per country, the maximum number of tokens per investor and any other necessary conditions for a particular asset. This verification is done by calling the function in the contract. Compliance canTransfer Compliance If all these conditions are met, then the**_transfer** function is executed, which transfers the tokens, then the function in the contract is called to update the state (what is executed in it depends on what conditions will be written there), and returns true. If the necessary conditions are not reached at any stage, the execution of this function will be reverted with an error. transferred Compliance Implementing regulatory rules directly into the code of smart contracts in this way can open up new opportunities by ensuring the legality and control of transactions by authorized entities. It is important to realize that most assets can only be tokenized if these conditions are met. For example, if we take securities that are recorded by government agencies, they have regulatory requirements and they cannot tokenize them without a guarantee of compliance with those requirements. If these things are not implemented, it makes it impossible for the regulated institutional players, which account for the vast majority of capital in the world, to participate. Because the institutional players are regulated, they cannot transact in assets that are not allowed for them. The government can allow, the government needs regulation, it's a vicious circle. You can have different attitudes to control and regulation, but right now there are a lot of pluses in this direction, which can really give a serious impulse to the growth of this direction at the expense of mass adoption and large players. Use Cases The ERC-3643 standard has been specifically designed for the regulated tokenization of Real World Assets. This solves many issues regarding assets that need to be regulated and therefore the ERC-3643 standard makes the following use cases possible: Tokenization of real commodities in the supply chain using ERC-3643 can increase transparency and efficiency of financing, especially by exactly following the terms and conditions that can be flexibly set in smart contracts, which also strongly reduces counterparty risk. Supply Chains: ERC-3643 can be used for the tokenization of private companies or startups, this increases accessibility to a wider range of investors and thus significantly increases liquidity in this area. Private companies and startups: Tokenization of real estate via the ERC-3643 standard not only increases liquidity for this asset class (especially expensive ones), but also makes transactions with them much more accessible and secure. This is due to the fact that we can set customized conditions for transactions, which will be verified directly in smart contracts. Real estate: ERC-3643 allows tokenization of any securities, including those backed by commodities (gold, oil etc.). This significantly increases their liquidity, simplifies tracking and control of compliance with the necessary requirements for transactions (due to the fact that we can set any requirements and restrictions). Also in the case of tokenized securities, deals are much faster and the data is immediately updated in the blockchain. Securities: By tokenizing diamonds and other treasures through the ERC-3643 standard, we are able to effectively control them and increase liquidity. Liquidity is increased by representing them on the open blockchain as tokens. And by splitting them into smaller pieces, it is possible to lower the entry threshold and increase liquidity even more. The standard is ERC-20 compatible. This means that these tokens can be traded on decentralized exchanges, which adds to diamonds and other treasures the liquidity of funds that are placed on the blockchain. Transactions will be easier to control because they will strictly follow the rules that will be set, and using the blockchain as a ledger makes it easier to track. Diamonds and other treasures: Conclusion ERC-3643 implements the needs of the majority of participants in traditional markets and makes it possible for them to move to the Web3 direction through tokenization. For them, it opens up opportunities to increase efficiency and improve processes. For the Web3 direction itself, it can mean a great inflow of funds and acceleration of development rates, which opens up a great expansion potential. Tokenization allows optimization and increases the efficiency of owning and circulating RWAs. These benefits include increased security of transactions, 24/7 availability, lower transaction costs, and increased liquidity. These are definitely benefits that are absolutely worth using, but before this can happen, a number of regulatory issues need to be resolved. To solve them, a standard for EVM networks (Ethereum) has been created, which allows for such regulation and makes mass adoption of tokenization of real word assets a step closer. The information presented in this article is the subjective opinion of the author and should not be considered a recommendation or a call to action. Disclaimer: The information presented in this article may be incomplete or not fully accurate and readers are advised to conduct their own research to confirm and supplement the information presented. Readers should understand that making any decisions based on this information involves risks, including loss of capital or other negative consequences. Readers are advised to seek professional advice and evaluate their own circumstances before making any decisions. The author shall not be liable for any loss or damage resulting from the use of information from this article.

Tokenizing Real-World Assets (RWA) Part 2: A Deeper Dive Into ERC-3643

This paper is available on arxiv under CC 4.0 license. Authors: (1) Eleonora Alei, ETH Zurich, Institute for Particle Physics & Astrophysics & National Center of Competence in Research PlanetS; (2) Björn S. Konrad, ETH Zurich, Institute for Particle Physics & Astrophysics & National Center of Competence in Research PlanetS; (3) Daniel Angerhausen, ETH Zurich, Institute for Particle Physics & Astrophysics, National Center of Competence in Research PlanetS & Blue Marble Space Institute of Science; (4) John Lee Grenfell, Department of Extrasolar Planets and Atmospheres (EPA), Institute for Planetary Research (PF), German Aerospace Centre (DLR) (5) Paul Mollière, Max-Planck-Institut für Astronomie; (6) Sascha P. Quanz, ETH Zurich, Institute for Particle Physics & Astrophysics & National Center of Competence in Research PlanetS; (7) Sarah Rugheimer, Department of Physics, University of Oxford; (8) Fabian Wunderlich, Department of Extrasolar Planets and Atmospheres (EPA), Institute for Planetary Research (PF), German Aerospace Centre (DLR); (9) LIFE collaboration, www.life-space-mission.com. Table of Links Abstract & Introduction Methods Results Discussion Conclusions Next Steps & References Appendix A: Scattering of terrestrial exoplanets Appendix B: Corner Plots Appendix C: Bayes’ factor analysis: other epochs Appendix D: Cloudy scenarios: additional figures Appendix C: Bayes’ factor analysis: other epochs As described in Section 2, the theoretical spectral model was updated with respect to Paper III and it now takes into account additional physical processes. For the results presented in Section 3, we ran retrievals using the most updated version of the Bayesian framework. The additional flexibility of petitRADTRANS now allows us to quantify the impact of CIA and scattering in retrievals. We ran additional retrievals on the clear sky scenarios for R = 50 and S/N = 10. In these retrievals, we altered the number of physical processes that were treated in the petitRADTRANS theoretical spectral model as follows: – Including both CIA and scattering (setup used in Section 3); – Excluding both CIA and scattering; – Including scattering and excluding CIA; – Including CIA and excluding scattering. In the runs where scattering is included, we consider selfscattering, surface scattering of the thermal radiation, and gaseous Rayleigh scattering (see Table 5 for references). We do not include aerosol and cloud scattering in the calculation. Since our theoretical spectral model neglects clouds, in this analysis we considered only the cloud-free scenarios. The effect of modeling cloudy spectra using a cloud-free retrieval model will be discussed in detail in the Section 4.2. To determine the theoretical spectral model configuration that best reproduces the input spectra we performed a Bayes’ factor analysis. The Bayes’ factor is defined as: The results for the remaining epochs exhibit similar behaviour. In Figure C.2, we show the results for the NOE Earth, Figure C.3 shows the ones for the GOE Earth, and the results for the prebiotic Earth are shown in Figure C.4. This paper is available on under CC 4.0 license. arxiv

Baye's Factor Analysis - Other Epochs: A Deeper Look Into Our Exoplanet Study

Stories, News & Mentions about QEEPER

Qeeper’s Stories on HackerNoon