Kaizan

kaizan

Startups of The Year 2023 Winners: Europe

Business owners who manufacture their own products encounter several challenges when it comes to managing inventory. This is because their data is often spread across multiple disparate systems that don’t connect to each other — resulting in countless hours of manual data entry and inconsistencies that can cause stock-outs. This is where Kaizntree comes in, an online platform that centralizes the inventory management process by consolidating all data into one interface. It also leverages artificial intelligence to predict demand trends and provide actionable insights for business growth.
Learn more about how Kaizntree is helping business owners around the world better manage their inventory. The Ripple Effect of Poor Inventory Tracking and the Inception of Kaizntree Business owners who manufacture their own products tend to operate across various disconnected platforms. So, while people often assume e-commerce websites like Shopify are all you need to manage a product business, the reality is that most e-commerce platforms only help you manage finished products — leaving business owners needing to track packaging, raw materials, and production elsewhere. As a result, most business owners lack a centralized platform to manage their inventory and logistics and often end up spending an average of 30 hours each month manually re-entering data across their sales channels, spreadsheets, and accounting software. This fragmented and tedious workflow is prone to human error and often results in decisions based on inaccurate data, such as incorrect stock counts, with reports showing that 60% of businesses have lost important data due to human error. These problems lead to more significant issues like not being able to fulfill customer orders and unexpectedly running out of important materials, eventually resulting in substantial lost revenue and a damaged brand reputation. That’s where Kaizntree comes in. NYU Shanghai graduates Marcos Brisson and Benoît Van Keer first developed the idea for Kaizntree while working as business consultants. As they built their client base, they noticed a lot of small business owners struggled with tracking stock and knowing how much to make of each product. According to Brisson, “We saw firsthand how business owners would pour everything into growing sales, then crumble under the weight of inventory problems.” The founders credited this issue to the fragmented ecosystem in which business owners operate and the lack of a simple solution that pulls it all together. There are various inventory platforms on the market that claim to be for small business owners, but “it’s as if these systems were built for accountants, not business owners,” Brisson notes. With this in mind, the pair developed Kaizntree to simplify the complexity of making your own product and selling it across multiple channels. It’s a user-friendly inventory software that serves as the single source of truth for businesses to oversee their operations — from raw materials to production and sales — via a centralized platform. How Kaizntree Automates Inventory Workflow What sets Kaizntree apart from other solutions is a focus on not only automating manual data entry but also providing businesses with actionable insights. For example, Kaizntree can forecast your sales demand and then tell you exactly what products need to be made and what raw materials need to be ordered. Another key point is Kaizntree’s streamlined onboarding process. All you need to do is upload your existing spreadsheets and connect your sales channels to Kaizntree, at which point the system’s AI automatically fills in the necessary details — eliminating the need for manual data entry. Once data has been successfully uploaded, Kaizntree proceeds to update your sales channels whenever products are manufactured or sold. It also tracks the costs of raw materials and sends alerts whenever stock levels are running low, allowing you to focus on growth without having to worry about running out of stock. Additionally, Kaizntree utilizes a built-in predictive algorithm that analyzes sales history to make data-driven suggestions on optimal stock levels. This helps you prevent future stock-outs at the finished product, raw material, or packaging level. You can also generate purchase orders for raw materials and send them to suppliers directly from the platform, which is essential to maintaining an efficient production workflow and scaling on demand. As Kaizntree continues to expand its automation features, you can rely on accurate data and AI-driven insights to enhance your workflow and successfully grow your operations. As Brisson puts it, “Our goal is to be the foundation on which business owners worldwide can start, scale, and manage their entire operations.” Business owners who manufacture their own products encounter several challenges when it comes to managing inventory. This is because their data is often spread across multiple disparate systems that don’t connect to each other — resulting in countless hours of manual data entry and inconsistencies that can cause stock-outs. This is where Kaizntree comes in, an online platform that centralizes the inventory management process by consolidating all data into one interface. It also leverages artificial intelligence to predict demand trends and provide actionable insights for business growth.
Learn more about how Kaizntree is helping business owners around the world better manage their inventory. The Ripple Effect of Poor Inventory Tracking and the Inception of Kaizntree Business owners who manufacture their own products tend to operate across various disconnected platforms. So, while people often assume e-commerce websites like Shopify are all you need to manage a product business, the reality is that most e-commerce platforms only help you manage finished products — leaving business owners needing to track packaging, raw materials, and production elsewhere. As a result, most business owners lack a centralized platform to manage their inventory and logistics and often end up spending an average of 30 hours each month manually re-entering data across their sales channels, spreadsheets, and accounting software. This fragmented and tedious workflow is prone to human error and often results in decisions based on inaccurate data, such as incorrect stock counts, with reports showing that 60% of businesses have lost important data due to human error. These problems lead to more significant issues like not being able to fulfill customer orders and unexpectedly running out of important materials, eventually resulting in substantial lost revenue and a damaged brand reputation. 60% of businesses have lost important data 60% of businesses have lost important data That’s where Kaizntree comes in. NYU Shanghai graduates Marcos Brisson and Benoît Van Keer first developed the idea for Kaizntree while working as business consultants. As they built their client base, they noticed a lot of small business owners struggled with tracking stock and knowing how much to make of each product. According to Brisson, “We saw firsthand how business owners would pour everything into growing sales, then crumble under the weight of inventory problems.” The founders credited this issue to the fragmented ecosystem in which business owners operate and the lack of a simple solution that pulls it all together. There are various inventory platforms on the market that claim to be for small business owners, but “it’s as if these systems were built for accountants, not business owners,” Brisson notes. With this in mind, the pair developed Kaizntree to simplify the complexity of making your own product and selling it across multiple channels. It’s a user-friendly inventory software that serves as the single source of truth for businesses to oversee their operations — from raw materials to production and sales — via a centralized platform. Kaizntree Kaizntree How Kaizntree Automates Inventory Workflow What sets Kaizntree apart from other solutions is a focus on not only automating manual data entry but also providing businesses with actionable insights. For example, Kaizntree can forecast your sales demand and then tell you exactly what products need to be made and what raw materials need to be ordered. Another key point is Kaizntree’s streamlined onboarding process. All you need to do is upload your existing spreadsheets and connect your sales channels to Kaizntree, at which point the system’s AI automatically fills in the necessary details — eliminating the need for manual data entry. Once data has been successfully uploaded, Kaizntree proceeds to update your sales channels whenever products are manufactured or sold. It also tracks the costs of raw materials and sends alerts whenever stock levels are running low, allowing you to focus on growth without having to worry about running out of stock. Additionally, Kaizntree utilizes a built-in predictive algorithm that analyzes sales history to make data-driven suggestions on optimal stock levels. This helps you prevent future stock-outs at the finished product, raw material, or packaging level. You can also generate purchase orders for raw materials and send them to suppliers directly from the platform, which is essential to maintaining an efficient production workflow and scaling on demand. As Kaizntree continues to expand its automation features, you can rely on accurate data and AI-driven insights to enhance your workflow and successfully grow your operations. As Brisson puts it, “Our goal is to be the foundation on which business owners worldwide can start, scale, and manage their entire operations.”

How Kaizntree Is Becoming the Operations Management Platform for Business Owners Around the World

কাইজন্ট্রি কীভাবে সারা বিশ্বের ব্যবসার মালিকদের জন্য অপারেশন ম্যানেজমেন্ট প্ল্যাটফর্ম হয়ে উঠছে

Wie Kaizntree zur Betriebsmanagement-Plattform für Unternehmer auf der ganzen Welt wird

Cómo Kaizntree se está convirtiendo en la plataforma de gestión de operaciones para propietarios de empresas de todo el mundo

Comment Kaizntree devient la plateforme de gestion des opérations pour les propriétaires d'entreprises du monde entier

कैसे Kaizntree दुनिया भर के व्यापार मालिकों के लिए संचालन प्रबंधन मंच बन रहा है

Kaizntree が世界中のビジネスオーナーのための運用管理プラットフォームになる方法

Kaizntree가 전 세계 사업주를 위한 운영 관리 플랫폼이 되는 방법

Como o Kaizntree está se tornando a plataforma de gerenciamento de operações para proprietários de empresas ao redor do mundo

Как Kaizntree становится платформой управления операциями для владельцев бизнеса по всему миру

Kaizntree, Dünya Çapındaki İşletme Sahipleri İçin Operasyon Yönetimi Platformu Nasıl Oluyor?

Kaizntree đang trở thành nền tảng quản lý hoạt động cho các chủ doanh nghiệp trên toàn thế giới như thế nào

Kaizntree 如何成为全球企业主的运营管理平台

Writing this personal review on how my year went is kind of emotional. I have always procrastinated writing or avoided it the previous years, thinking I was no match to others writing theirs. I am a lot courageous now, and I am grateful for how the journey so far has been for me in 2021. Choosing the title with the word kaizen was something I learned from from his "Future Proof Yourself" talk earlier in the year on Zero To Mastery (ZTM). Andrei Neagoie What is kaizen? For me, a simple explanation of what kaizen means is the continuous slight incremental improvement I have witnessed in 2021. Kaizen is a Sino-Japanese word. Work This year, I began the year jobless, and it was a tough one, and my mental health wasn't all that great. I persevered and made a promise that I would make it no matter how little my win came.
Here are some of my highlights that filled my face with a smile. I became an author with and participated as a technical editor during the content creation hackathon in partnership with and . Hackmamba Auth0 Cloudinary Got a job as a full-stack developer in an ed-tech company in Lagos. Got a badge from my participation in the HarperDB hackathon for my job board application. Taught a group of enthusiastic web developers as a web development instructor even when I doubted myself if I could do it. I did pretty well for the 7 months I partook in the exercise. Made a couple of six figures in NGN from content creation. I became happy, and it stabilised my finances. The next target is 7 figures and higher in NGN. On October 1, 2021, I received an email that I got in to be AWS CB (Community Builder) which came as a shock because I had forgotten I ever applied for the program. Next was congratulatory email on becoming a blogging fellow. 2022 will be filled with me writing a lot. HackerNoon Networking and connecting with people virtually made me calmer and relieved as I have worked most of the year from home. Even though I promised myself I would speak at tech conferences this year, fear and imposter syndrome always kicked in 😜. Next thing you guessed right, I will say next time I send my proposal to speak. But finally, I gave a talk in December 2021 on Clubhouse. Speaking engagement The not so pleasant ride Remarkable events occurred in 2021 but with a twist of rejections, failure, and ways to improve in 2022. Got rejected for a lot of open roles. About 60+ or more rejections. Some never got back with "unfortunately" 🤐 Received an L in August as I left my full-stack developer role Failed at the last stage of the MEST EIT and didn't get the chance to be in Ghana 😭 Technical writing saved me from my depression, and I hope to get better at it and churn out more content Moving forward Going into 2022, I will be more intentional about what I want to achieve. Try to take fitness more seriously and be stronger Learn the craze and buzz about Web 3.0 Also, I want to learn about the standard practice of DevOps Make more money 💰. I am not shy about stating my intention Conclusion As mentioned in the work section of my review, some events in 2021 made me smile, and I promise to build on that and go harder in 2022 and make more memories that weren't as pleasant in 2021. Things can only get better from now on as this write-up has been on my mind all year, and I finally inked it on the digital space (the internet) 🚀. Finally, 1% improvement every day is what is categorising my 2022. And I believe so much in the kaizen process. If you liked this article, let's connect deeper on | . Twitter LinkedIn Also published on: https://dev.to/terieyenike/2021-kaizen-or-nothing-2c57

2021: Reviewing and Kaizen-ing My Programming and Writing Life

HackerNoon Reporter: Please tell us briefly about your  background. I'm from New Zealand where I studied business and languages at university - in fact studied Japanese and Chinese, and even chose to study in Japan for a year. Since then I've worked in startups, large consultancies and technology companies across the UK, Europe and New Zealand. My previous experience includes solution architecture, product management and management consulting. What's your startup called? And in a sentence or two, what does it do? Kaizen Languages teaches learners how to speak a language by giving you your own AI tutors to practice speaking with. Our AI tutors help you experiment in your new language and correct your pronunciation, grammar and spelling. What is the origin story? Patrick (CTO & Cofounder) and I were talking about the languages we had learned and how difficult it was to acquire languages. Doing more research, we found the biggest blocker for most language learners is simply the lack of speaking practice. To learn a language like Spanish it takes 600 hours of speaking practice, and getting that practice is incredibly difficult if you don't live in a Spanish speaking country! In 2018, we decided to create a prototype that used AI conversations as the basis for teaching and practicing languages. The beta was so successful that we raised capital and now have a team dedicated to helping learners reach fluency. What do you love about your team, and why are you the ones to solve this problem? The Kaizen Languages team are a bunch of linguistic enthusiasts. Everyone can speak multiple languages and we all love learning more about how each other's languages work and are taught. We're also obsessed with tech and how powerful it can be in helping others achieve their goals. But fundamentally, we're obsessed with the mission -  making language learning easier than ever. If you weren’t building your startup, what would you be doing? I would probably be building another startup, like many entrepreneurs I have many ideas on great new products and am addicted to finding ways to make them real. At the moment, how do you measure success? What are your core metrics? We measure success in how many lessons learners complete and how many new concepts they can learn in life-like situations. Language learning isn’t a sprint and those who succeed know that keeping up the habit is half the pattern. Knowing this, we constantly encourage users to continue learning by completing lessons. What’s most exciting about your traction to date? We've had over 300k downloads to date whilst maintaining a 4.8 star rating, and that's just teaching Japanese! We recently launched our Android app, as well as a Spanish language course that's off to a promising start. What technologies are you currently most excited about, and most worried about? And why? AI is a very exciting space and seeing what is constantly evolving in the space is incredible. It's also a very broad space with many breakthroughs in areas outside of just NLP. What worries me is with any new technology is how it can be used without consideration of its impacts. AI is an area where it has tremendous potential, but this can be both positive and negative. What drew you to get published on HackerNoon? What do you like most about our platform? The depth of content covering other startups, founders but also the educational aspect of different tech skills. Kaizen Languages is nominated for a , UK Startup of the year in London

AI's Role in Language Learning: Stuart Barrass, Kaizen Languages CEO

- When nothing seems to help, I go and look at a stonecutter hammering away at his rock, perhaps a hundred times without as much as a crack showing in it. Yet at the hundred and first blow, it will split in two, and I know it was not that last blow that did it—but all that had gone before. Jacob A. Riis How are post-war industries rebuilt? Can worldwide poverty be reduced? Can couples' divorce rates be predicted? The answer to such dramatic questions is usually innovation. Innovation is a revolutionary process that requires an entirely different mindset to find novel solutions. And yet, innovation is not the only strategy to create dramatic change. If innovation happens quickly, requiring shocking transition and radical adaption, kaizen is an approach to commit to small and comfortable steps towards better circumstances. This (literal translation is “change for better”) is American in origin. It was brought initially to Japan by American statisticians and quality experts such as W. Edwards Deming, who helped Japan rebuild its industry after World War II. Toyota made kaizen famous by adopting the lean manufacturing process and the iterative improvement mindset. So, in the 1980s, kaizen crossed the world back to the United States, mainly in highly technical business settings. Japanese concept Nevertheless, kaizen is also a good strategy for personal self-improvement. Too many times, we tend to overlook the power of small steps. Still, over extended periods, tiny steps will compound into significant results. Instead of focusing on daunting questions and high expectations, we can break these targets into absurdly little milestones that we will try to hit consistently. The kaizen framework The kaizen process looks like this: First, we choose a novel habit that we want to incorporate into our daily routines or an action that makes us feel uncomfortable. Then, we repeatedly ask small questions on how to break this action into small steps. Following insightful answers, we break that activity into tiny little steps. We allocate low time unit commitments (varying from seconds to minutes or tens of minutes) to fulfill them every day easily. Accountability is critical as it helps with setting clear intentions and keeping a straight course. Examples: keeping a journal, reviewing the day in a stoic manner, or having a partner, like a friend, online buddy, psychologist, etc. The goal is to maintain these new activities effortlessly. There are subtle psychological nuances about making the first step. After experiencing the first step, we can decide if it is time to make another. We will know when we are ready, as this process will become automatic. But, if at any level, we find ourselves dreading the activity or making excuses to get rid of it, psychologist Robert Mauer suggests in his book , acknowledging that it is time to reduce the size of the step. After all, life is not a game of perpetual ladders but more a game of snakes and ladders. One Small Step Can Change Your Life: The Kaizen Way Why is repetition necessary? Because repetition does not make it perfect but permanent, whenever we repeat a new behavior, even for just a few seconds, eventually, the brain will stamp those actions as meaningful. As a result, it will create new neural pathways to reinforce those new behaviors. Incremental and bite-size adjustments have a water-like quality: give them enough time, intention, and focus, and changes will start to permeate our lives like water absorbed by dry soil. Why is it essential to ask small questions? Sometimes we are so concerned about failure that we deliberately create vague goals, and thus, nobody can criticize us when we crash. Perhaps it helps to break the impossible questions into more manageable chunks. Asking small questions might also lead to more truthful answers. Focusing on asking questions allows the brain to concentrate on problem-solving. By repeating this behavior long enough, the brain will eventually lead us to valuable and novel insights or exciting approaches to our questions. Why is it important to take small actions? The brain tends to resist change and favors the sweet equilibrium of . But small actions do not look perilous, so we rewire our nervous system by employing tiny steps. Thus, it is possible to hit a new homeostatic state where we fully incorporated and maintain the new habits. homeostasis Examples of kaizen questions What is one small step I could take toward reaching my goal? What is one small step I could take to improve my health (or relationships, career, or any other area)? If you tend to berate yourself with negative questions (Why am I so fat?), try asking: What is one thing I like about myself today? If health were my first priority, what would I be doing differently today? What is one way I can remind myself to drink more water? How could I incorporate a few more minutes of exercise into my daily routine? If you are unhappy but aren’t sure why, try asking yourself this: If I were guaranteed not to fail, what would I be doing differently? This question is for anyone who has a festering conflict with another person, whether a boss, employee, in-law, or neighbor, and is trying to get past this problem. Every day, ask yourself: What’s one good thing about this person? - Robert Maurer, One Small Step Can Change Your Life: The Kaizen Way Examples of kaizen actions – Remove one object from the shopping cart before heading to the cash register. Stop overspending – Stand—yes, just stand!—on the treadmill for a few minutes every morning. Begin an exercise program – Once a day, note where your body is holding tension (your neck? lower back? shoulders?). Then take one deep breath. Manage stress – Pick an area of the house, set a timer for five minutes, and tidy up. Stop when the timer goes off. Keep the house clean - Robert Maurer, One Small Step Can Change Your Life: The Kaizen Way Simply standing on the treadmill will not produce much health. But such tiny changes can bring winds of attitude change as initial resistance starts melting. Mauer writes of his patient, Julie, that she became more animated after standing on the treadmill for a minute daily and wanted to know what else can do on a few minutes per day. Within a few months, Julie found herself exercising even when she didn’t have to. The exponential growth of kaizen habits Usually, to explain exponential growth, teachers use the water lily problem: consider a lake with just one lily pad on a summer day of June 1. Now consider that every day lily pads will double in numbers. By June 30, lily pads will cover the whole lake. Q: When was 50% of the lake covered? A: June 29. As lily pads double in size, the lily pads will go from 50% on June 29 to 100% on June 30. Q: If lily pads cover the lake 100% on June 30 and 50% on June 29, when did the lily pads cover 1% of the lake? A: June 24th. (June 30 – 100%, June 29 – 50%, June 28 – 25%, June 27 – 12%, June 26 – 6%, June 25 – 3%, June 24 – 1%) It took more than three weeks for lily pads to cover just a tiny 1% of the lake and then less than a week to cover the whole lake. Similarly, there might be years behind an “overnight” success as it can take a long time to build a strong foundation and a much shorter time to get from 50% to 100% with kaizen. The negative side of kaizen As this shows, we might be chasing obsessive optimizations. Let’s ask ourselves: are all the improvements we want to apply worth our time, energy, and focus? timeless xkcd cartoon Then, as the saying goes, it can’t be a superpower if it can’t be used for evil. The beginning of addiction usually follows kaizen. Just a drink, just a cigarette, just a pill, something so tiny, so small we would think it would not matter at all. Except that it does, and so, proceed with caution. The marginal gains of kaizen and relieving global poverty In 2019, Esther Duflo, her husband Abhijit Banerjee, and Michael Kramer shared the Nobel Memorial Prize in Economic Sciences “for their experimental approach to “. Their approach to fighting global poverty is based on scientific evidence. They ran hundreds of (trials where participants are randomly allocated to various treatments) to find which approach yields better results. alleviating global poverty randomized controlled trials Scientifically approaching kaizen does not mean making small changes and hope for the best. Instead, it is about breaking a complex problem into small parts. Then, by using empirical evidence from controlled and randomized experimentation, effects are isolated. Thus, it can be established which steps work and which don’t. For example, they discovered that: The study of the long-term effect of deworming children in Kenya, mentioned above, concluded that being dewormed for two years instead of one (and hence being better nourished for two years instead of one) would lead to a lifetime income gain of $3,269 USD PPP. Small differences in investments in childhood nutrition (in Kenya, deworming costs $1.36 USD PPP per year; in India, a packet of iodized salt sells for $0.62 USD PPP; in Indonesia, fortified fish sauce costs $7 USD PPP per year) make a huge difference later on. As Duflo declared to author Matthew Syed: It is very easy to sit back and come up with grand theories about how to change the world. But often, our intuitions are wrong. The world is too complex to figure everything out from your armchair. The only way to be sure is to go out and test your ideas and programs and to realize that you will often be wrong. But that is not a bad thing. It leads to progress. It is possible to make significant progress against the biggest problem in the world through the accumulation of a set of small steps, each well thought out, carefully tested, and judiciously implemented. -Matthew Syed, Black Box Thinking: Why Most People Never Learn from Their Mistakes–But Some Do Kaizen and predicting divorce rates John Gottman is an American psychological researcher who did over 40 years of extensive research on divorce prediction. Analyzing thousands of marriages at their couples laboratory , Gottman has been able to predict with high accuracy ( ) if couples would be happily married or divorced within four years. Love Lab based on experiments, varying from 87% to 94% accuracy One of the groundbreaking findings was that positive remarks outweigh negative remarks with a daily factor of five to one in successful relationships. Long-lasting marriages weren’t about over-the-top grand gestures such as expensive holidays or memorable birthday parties, but they were more about the quiet, almost invisible gestures. Partners were asking each other about their days. They used a pleased tone of voice when the partner called instead of an exasperated or rushed voice. Small gestures of affection are more predictive of loving and trusting relationships, presumably because repeated gentle moments provide consistent nourishing. And as for the name of ? Gottman’s podcast Small things often. This reminds me of what I read in : Maybe You Should Talk to Someone One woman said to her husband, “You know what three words are even more romantic to me than ”I love you”?” “You look beautiful?” he tried. “No,” his wife said. “I understand you.” Kaizen is deceptively simple and utterly brilliant. It would take a lot of focus, failures, regrouping, and trust in the system, not results. To continue working and tinkering on imperceptible changes day after day, year after year, with no 100% guarantee the road we take will lead anywhere. It doesn’t matter. Samuel Beckett Try again. Fail again. Fail better. - . Previously published at https://www.roxanamurariu.com/the-exponential-art-of-kaizen/

Change for Better: The Exponential Art of Kaizen

Part 5 of The where I interview my heroes. series This is another very special version of the series. Today, I’m honored to be talking to Kaggle Competitions Grandmaster (Ranked#3), Discussions Master (Rank#5). Dr. Marios Michailidis: About the Series: I have very recently started making some progress with my . But to be honest, it wouldn’t be possible at all without the amazing community online and the great people that have helped me. In this Series of Blog Posts, I talk with People that have really inspired me. Self-Taught Machine Learning Journey The motivation behind doing this is, you might see some patterns and hopefully you’d be able to learn from the amazing people that I have had the chance of from. learning **Sanyam Bhutani:**​ Hello Grandmaster, Thank you for taking the time to do this. No problem. Thank you for considering me for this one :) Dr. Marios Michailidis: **Sanyam Bhutani:**​ You’re on the Top 3 of the Competitions leaderboard today. Could you tell the readers about your kaggle journey. What got you interested in competing on kaggle and about your path to becoming #1 in 2016? I went to study at Southampton University (to do a Master in Risk Management). Near the end of the last semester, I started joining some entrepreneurship talks in order to get some ideas about what to do next. There was a specific talk about an entrepreneur who described his journey of making a successful business out of horse racing predictions using Logistic regression. I thought it was quite impressive as he basically started with pretty much nothing and ended up making a good living out of it. Then I felt I wanted to have the same “power” to predict the future and I joined the field of predictive modelling. Later on, I joined kaggle too. Dr. Marios Michailidis: You’re a Competitive Data Scientist and also an open source contributor to projects: StackNet, KazAnova. Sanyam Bhutani: Where does kaggle come in the picture? Is it related to your other projects and the research work? Kaggle helps me in various ways: Dr. Marios Michailidis: Learn new skills, new , what’s hot. tools Solve a variety of problems. Become part of a (very generous) and open community. Collaborate with other experienced people in the field. Test/prove my ideas. Benchmark myself and how well I can do in a variety of problems. Receive recognition, promote my research. Do my job better. The company I work for: , is a leader in the development of software for data science and predictive analytics. Kaggle is a great environment to test our products and ensure they fair well against some of the top data scientists in the field. H2o.ai Generally become better in my craft. You’ve had many amazing finishes at many competitions.Could you tell what was your favourite challenge? Sanyam Bhutani: It was the first competition I won. I loved working with my then-teammate Gert. Plus the concept of the competition was very close to what I was working on at that point (in recommenders) and my Ph.D. Dr. Marios Michailidis: — -The Acquire valued shoppers ’ challenge, What kind of challenges do you look for today? How do you decide to enter a new competition? Sanyam Bhutani: I enter all**!** I think the question here should be which challenges I choose NOT to enter! The only challenges I will not enter are those that are very demanding resource-wise (e.g. some of the computer vision ones with close to a TB of data). Other than that I am up for any challenge! Dr. Marios Michailidis: What are your first steps and go to techniques when starting out on a new competition? Sanyam Bhutani: Dr. Marios Michailidis: Understand the problem and the we are tested on- this is key. metric Create a reliable cross-validation process that best would resemble the leaderboard or the test set in general as this will allow me to explore many different algorithms and approaches, knowing the impact they could yield. Understand the importance of different algorithmic families, to see when and where to maximize the intensity (is it a linear or non-linear type of problem?) Try many different approaches/techniques on the given problem and seize it from all possible angles in terms of algorithms ‘selection, hyper parameter optimization, feature engineering, missing values’ treatment- I treat all these elements as hyperparameters of the final solution. For the readers and noobs like me who want to become better kagglers, what would be your best advice? Sanyam Bhutani: Dr. Marios Michailidis: Dedicate time to learning about data science and actually doing “kaggling”. You will eventually see results, sooner or later. You don’t need a Ph.D. to do this. Don’t be intimidated by others’ scores. We all started from the bottom (and occasionally we still find ourselves near there on different problems)! Pick up some to enhance your skills. online coursera classes Also have a read at this: https://www.linkedin.com/pulse/how-start-data-science-marios-michailidis Given the explosive growth rate of ML, How do you stay updated with the recent developments? Sanyam Bhutani: Dr. Marios Michailidis: Kaggle Reading relevant blogs/data science pages Attending conferences Through networking What are your thoughts about Machine Learning as a field, do think its Overhyped? Sanyam Bhutani: A bit**.** In the sense that everyone thinks they should somehow apply/use machine learning without exactly knowing what it does or how it can help them ( or even if the really need it — recently I saw a toaster that was using machine learning — god knows for what reason!). However, I believe this field will flourish more in the following years. Dr. Marios Michailidis: Before we conclude, any tips for the beginners who aspire to be like you someday but feel completely overwhelmed to even start competing? Sanyam Bhutani: On top of what I added before regarding becoming better on kaggle Dr. Marios Michailidis: Let me rehearse what I had said in a previous interview on what wins kaggle competitions : Understand the problem well. Discipline; To have a well-thorough and documented approach that you follow strictly and defines all the modelling process/framework from how you cross-validate, select models, avoids overfitting (which requires a lot of …discipline). Allow room to try problem-specific things or new approaches within that framework. For example, unless you use deep learning for image classification, you are not likely to go very far. The hours you put in. Have access to the right tools. Good hardware. Make key partnerships. Look for people that are likely to have taken a very diverse approach from you. Ensembling (as in combing many different algorithms/approaches to get a better score). Thank you so much for doing this interview. Sanyam Bhutani: is the best community for kaggle where you can find Dr. Michailidis, Other Kaggle Grandmasters, Masters, Experts and it’s a community where even noobs like me are welcome. Kaggle Noobs Come join if you’re interested in ML/DL/Kaggle. If you found this interesting and would like to be a part of , you can find me on Twitter . My Learning Path here

Interview with Kaggle Competitions Grandmaster: KazAnova (Rank #3): Dr. Marios Michailidis

Authors:
(1) Tinghui Ouyang, National Institute of Informatics, Japan (thouyang@nii.ac.jp);
(2) Isao Echizen, National Institute of Informatics, Japan (iechizen@nii.ac.jp);
(3) Yoshiki Seo, Digital Architecture Research Center, National Institute of Advanced Industrial Science and Technology, Japan (y.seo@aist.go.jp). Table of Links Abstract and Introduction Description and Related Work of OOD Detection Methodology Experiments and Results Conclusions, Acknowledgement and References V. CONCLUSIONS In this paper, we addressed the issue of out-of-distribution (OOD) data in AI quality management by proposing a framework that combines deep learning and statistical measures for OOD detection. Initially, we leveraged the strong feature representation and dimensionality reduction capabilities of AE to extract activation traces from hidden neurons as input features for OOD analysis. Subsequently, we employed five statistical measures, namely KD, LOF, MD, kNN, and the proposed LCP, using these representative features for OOD detection. Our findings revealed that the proposed LCP, which incorporates both neighbor information and data reconstruction error, outperforms the other measures in OOD detection. Thus, it was proved to be valuable for describing and detecting OOD data. Furthermore, this research extracted reasonable corner case data with high OOD scores from the given datasets, such as MNIST, CIFAR10, and GTSRB datasets. These corner case data have high OOD scores and exhibit abnormal characteristics compared to normal data. Therefore, detecting such data using the proposed method is helpful in future AI quality assurance, particularly for quality analysis related to data security. VI. ACKNOWLEDGEMENT This research is supported by the New Energy and Industrial Technology Development Organization (NEDO) project ’JPNP20006’, and JSPS Grant-in-Aid for Early-Career Scientists (Grant Number 22K17961). REFERENCES [1] The National Institute of Advanced Industrial Science and Technology (AIST). Machine learning quality management guideline, 2023. https://www.digiarc.aist.go.jp/en/publication/aiqm/. [2] Kexin Pei, Yinzhi Cao, Junfeng Yang, and Suman Jana. Deepxplore: Automated whitebox testing of deep learning systems. In proceedings of the 26th Symposium on Operating Systems Principles, pages 1–18, 2017. [3] Tinghui Ouyang, Yoshinao Isobe, Saima Sultana, Yoshiki Seo, and Yutaka Oiwa. Autonomous driving quality assurance with data uncertainty analysis. In 2022 International Joint Conference on Neural Networks (IJCNN), pages 1–7. IEEE, 2022. [4] Guansong Pang, Jundong Li, Anton van den Hengel, Longbing Cao, and Thomas G Dietterich. Anomaly and novelty detection, explanation, and accommodation (andea). In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 4145–4146, 2021. [5] Jingkang Yang, Kaiyang Zhou, Yixuan Li, and Ziwei Liu. Generalized out-of-distribution detection: A survey. arXiv preprint arXiv:2110.11334, 2021. [6] Varun Chandola, Arindam Banerjee, and Vipin Kumar. Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3):1–58, 2009. [7] Marco AF Pimentel, David A Clifton, Lei Clifton, and Lionel Tarassenko. A review of novelty detection. Signal processing, 99:215–249, 2014. [8] Chuanxing Geng, Sheng-jun Huang, and Songcan Chen. Recent advances in open set recognition: A survey. IEEE transactions on pattern analysis and machine intelligence, 43(10):3614–3631, 2020. [9] R´emi Domingues, Maurizio Filippone, Pietro Michiardi, and Jihane Zouaoui. A comparative evaluation of outlier detection algorithms: Experiments and analyses. Pattern recognition, 74:406–421, 2018. [10] Jing Li, Pengbo Lv, Huijun Li, and Wanghu Chen. Outlier detection based on stacked autoencoder and gaussian mixture model. In 2022 IEEE International Conference on Big Data (Big Data), pages 3763–3769. IEEE, 2022. [11] Michael E Tipping and Christopher M Bishop. Probabilistic principal component analysis. Journal of the Royal Statistical Society Series B: Statistical Methodology, 61(3):611–622, 1999. [12] John A Quinn and Masashi Sugiyama. A least-squares approach to anomaly detection in static and sequential data. Pattern Recognition Letters, 40:36–40, 2014. [13] Reuben Feinman, Ryan R Curtin, Saurabh Shintre, and Andrew B Gardner. Detecting adversarial samples from artifacts. arXiv preprint arXiv:1703.00410, 2017. [14] Kimin Lee, Kibok Lee, Honglak Lee, and Jinwoo Shin. A simple unified framework for detecting out-of-distribution samples and adversarial attacks. Advances in neural information processing systems, 31, 2018. [15] Kathrin Grosse, Praveen Manoharan, Nicolas Papernot, Michael Backes, and Patrick McDaniel. On the (statistical) detection of adversarial examples. arXiv preprint arXiv:1702.06280, 2017. [16] Ahmed Abusnaina, Yuhang Wu, Sunpreet Arora, Yizhen Wang, Fei Wang, Hao Yang, and David Mohaisen. Adversarial example detection using latent neighborhood graph. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 7687–7696, 2021. [17] Bruno Andriamanalimanana, Ali Tekeoglu, Korkut Bekiroglu, Saumendra Sengupta, Chen-Fu Chiang, Michael Reale, and Jorge Novillo. Symmetric kullback-leibler divergence of softmaxed distributions for anomaly scores. In 2019 IEEE Conference on Communications and Network Security (CNS), pages 1–6. IEEE, 2019. [18] Jianbo Yu, Xiaoyun Zheng, and Shijin Wang. A deep autoencoder feature learning method for process pattern recognition. Journal of Process Control, 79:1–15, 2019. [19] Lukas Ruff, Robert Vandermeulen, Nico Goernitz, Lucas Deecke, Shoaib Ahmed Siddiqui, Alexander Binder, Emmanuel M¨uller, and Marius Kloft. Deep one-class classification. In International conference on machine learning, pages 4393–4402. PMLR, 2018. [20] Jinhan Kim, Robert Feldt, and Shin Yoo. Guiding deep learning system testing using surprise adequacy. In 2019 IEEE/ACM 41st International Conference on Software Engineering (ICSE), pages 1039–1049. IEEE, 2019. [21] Zenan Li, Xiaoxing Ma, Chang Xu, and Chun Cao. Structural coverage criteria for neural networks could be misleading. In 2019 IEEE/ACM 41st International Conference on Software Engineering: New Ideas and Emerging Results (ICSE-NIER), pages 89–92. IEEE, 2019. [22] Shiqing Ma, Yingqi Liu, Guanhong Tao, Wen-Chuan Lee, and Xiangyu Zhang. Nic: Detecting adversarial samples with neural network invariant checking. In 26th Annual Network And Distributed System Security Symposium (NDSS 2019). Internet Soc, 2019. [23] Tarem Ahmed. Online anomaly detection using kde. In GLOBECOM 2009-2009 IEEE Global Telecommunications Conference, pages 1–8. IEEE, 2009. [24] Siying Xu, Huiyi Liu, Liting Duan, andWenjingWu. An improved lof outlier detection algorithm. In 2021 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), pages 113–117. IEEE, 2021. [25] Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of machine learning research, 9(11), 2008. [26] Yann LeCun. The mnist database of handwritten digits. http://yann. lecun. com/exdb/mnist/, 1998. [27] Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny images. 2009. [28] Tom Fawcett. An introduction to roc analysis. Pattern recognition letters, 27(8):861–874, 2006. [29] Tinghui Ouyang, Vicent Sanz Marco, Yoshinao Isobe, Hideki Asoh, Yutaka Oiwa, and Yoshiki Seo. Corner case data description and detection. In 2021 IEEE/ACM 1st Workshop on AI Engineering-Software Engineering for AI (WAIN), pages 19–26. IEEE, 2021. [30] Stallkamp, Johannes, Marc Schlipsing, Jan Salmen, and Christian Igel. The German traffic sign recognition benchmark: a multi-class classification competition. In The 2011 international joint conference on neural networks, pages 1453-1460. IEEE, 2011. This paper is available on arxiv under CC 4.0 license. Authors: (1) Tinghui Ouyang, National Institute of Informatics, Japan (thouyang@nii.ac.jp); (2) Isao Echizen, National Institute of Informatics, Japan (iechizen@nii.ac.jp); (3) Yoshiki Seo, Digital Architecture Research Center, National Institute of Advanced Industrial Science and Technology, Japan (y.seo@aist.go.jp). Authors: Authors: (1) Tinghui Ouyang, National Institute of Informatics, Japan (thouyang@nii.ac.jp); (2) Isao Echizen, National Institute of Informatics, Japan (iechizen@nii.ac.jp); (3) Yoshiki Seo, Digital Architecture Research Center, National Institute of Advanced Industrial Science and Technology, Japan (y.seo@aist.go.jp). Table of Links Abstract and Introduction Abstract and Introduction Description and Related Work of OOD Detection Description and Related Work of OOD Detection Methodology Methodology Experiments and Results Experiments and Results Conclusions, Acknowledgement and References Conclusions, Acknowledgement and References V. CONCLUSIONS In this paper, we addressed the issue of out-of-distribution (OOD) data in AI quality management by proposing a framework that combines deep learning and statistical measures for OOD detection. Initially, we leveraged the strong feature representation and dimensionality reduction capabilities of AE to extract activation traces from hidden neurons as input features for OOD analysis. Subsequently, we employed five statistical measures, namely KD, LOF, MD, kNN, and the proposed LCP, using these representative features for OOD detection. Our findings revealed that the proposed LCP, which incorporates both neighbor information and data reconstruction error, outperforms the other measures in OOD detection. Thus, it was proved to be valuable for describing and detecting OOD data. Furthermore, this research extracted reasonable corner case data with high OOD scores from the given datasets, such as MNIST, CIFAR10, and GTSRB datasets. These corner case data have high OOD scores and exhibit abnormal characteristics compared to normal data. Therefore, detecting such data using the proposed method is helpful in future AI quality assurance, particularly for quality analysis related to data security. VI. ACKNOWLEDGEMENT This research is supported by the New Energy and Industrial Technology Development Organization (NEDO) project ’JPNP20006’, and JSPS Grant-in-Aid for Early-Career Scientists (Grant Number 22K17961). REFERENCES [1] The National Institute of Advanced Industrial Science and Technology (AIST). Machine learning quality management guideline, 2023. https://www.digiarc.aist.go.jp/en/publication/aiqm/. [2] Kexin Pei, Yinzhi Cao, Junfeng Yang, and Suman Jana. Deepxplore: Automated whitebox testing of deep learning systems. In proceedings of the 26th Symposium on Operating Systems Principles, pages 1–18, 2017. [3] Tinghui Ouyang, Yoshinao Isobe, Saima Sultana, Yoshiki Seo, and Yutaka Oiwa. Autonomous driving quality assurance with data uncertainty analysis. In 2022 International Joint Conference on Neural Networks (IJCNN), pages 1–7. IEEE, 2022. [4] Guansong Pang, Jundong Li, Anton van den Hengel, Longbing Cao, and Thomas G Dietterich. Anomaly and novelty detection, explanation, and accommodation (andea). In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 4145–4146, 2021. [5] Jingkang Yang, Kaiyang Zhou, Yixuan Li, and Ziwei Liu. Generalized out-of-distribution detection: A survey. arXiv preprint arXiv:2110.11334, 2021. [6] Varun Chandola, Arindam Banerjee, and Vipin Kumar. Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3):1–58, 2009. [7] Marco AF Pimentel, David A Clifton, Lei Clifton, and Lionel Tarassenko. A review of novelty detection. Signal processing, 99:215–249, 2014. [8] Chuanxing Geng, Sheng-jun Huang, and Songcan Chen. Recent advances in open set recognition: A survey. IEEE transactions on pattern analysis and machine intelligence, 43(10):3614–3631, 2020. [9] R´emi Domingues, Maurizio Filippone, Pietro Michiardi, and Jihane Zouaoui. A comparative evaluation of outlier detection algorithms: Experiments and analyses. Pattern recognition, 74:406–421, 2018. [10] Jing Li, Pengbo Lv, Huijun Li, and Wanghu Chen. Outlier detection based on stacked autoencoder and gaussian mixture model. In 2022 IEEE International Conference on Big Data (Big Data), pages 3763–3769. IEEE, 2022. [11] Michael E Tipping and Christopher M Bishop. Probabilistic principal component analysis. Journal of the Royal Statistical Society Series B: Statistical Methodology, 61(3):611–622, 1999. [12] John A Quinn and Masashi Sugiyama. A least-squares approach to anomaly detection in static and sequential data. Pattern Recognition Letters, 40:36–40, 2014. [13] Reuben Feinman, Ryan R Curtin, Saurabh Shintre, and Andrew B Gardner. Detecting adversarial samples from artifacts. arXiv preprint arXiv:1703.00410, 2017. [14] Kimin Lee, Kibok Lee, Honglak Lee, and Jinwoo Shin. A simple unified framework for detecting out-of-distribution samples and adversarial attacks. Advances in neural information processing systems, 31, 2018. [15] Kathrin Grosse, Praveen Manoharan, Nicolas Papernot, Michael Backes, and Patrick McDaniel. On the (statistical) detection of adversarial examples. arXiv preprint arXiv:1702.06280, 2017. [16] Ahmed Abusnaina, Yuhang Wu, Sunpreet Arora, Yizhen Wang, Fei Wang, Hao Yang, and David Mohaisen. Adversarial example detection using latent neighborhood graph. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 7687–7696, 2021. [17] Bruno Andriamanalimanana, Ali Tekeoglu, Korkut Bekiroglu, Saumendra Sengupta, Chen-Fu Chiang, Michael Reale, and Jorge Novillo. Symmetric kullback-leibler divergence of softmaxed distributions for anomaly scores. In 2019 IEEE Conference on Communications and Network Security (CNS), pages 1–6. IEEE, 2019. [18] Jianbo Yu, Xiaoyun Zheng, and Shijin Wang. A deep autoencoder feature learning method for process pattern recognition. Journal of Process Control, 79:1–15, 2019. [19] Lukas Ruff, Robert Vandermeulen, Nico Goernitz, Lucas Deecke, Shoaib Ahmed Siddiqui, Alexander Binder, Emmanuel M¨uller, and Marius Kloft. Deep one-class classification. In International conference on machine learning, pages 4393–4402. PMLR, 2018. [20] Jinhan Kim, Robert Feldt, and Shin Yoo. Guiding deep learning system testing using surprise adequacy. In 2019 IEEE/ACM 41st International Conference on Software Engineering (ICSE), pages 1039–1049. IEEE, 2019. [21] Zenan Li, Xiaoxing Ma, Chang Xu, and Chun Cao. Structural coverage criteria for neural networks could be misleading. In 2019 IEEE/ACM 41st International Conference on Software Engineering: New Ideas and Emerging Results (ICSE-NIER), pages 89–92. IEEE, 2019. [22] Shiqing Ma, Yingqi Liu, Guanhong Tao, Wen-Chuan Lee, and Xiangyu Zhang. Nic: Detecting adversarial samples with neural network invariant checking. In 26th Annual Network And Distributed System Security Symposium (NDSS 2019). Internet Soc, 2019. [23] Tarem Ahmed. Online anomaly detection using kde. In GLOBECOM 2009-2009 IEEE Global Telecommunications Conference, pages 1–8. IEEE, 2009. [24] Siying Xu, Huiyi Liu, Liting Duan, andWenjingWu. An improved lof outlier detection algorithm. In 2021 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), pages 113–117. IEEE, 2021. [25] Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of machine learning research, 9(11), 2008. [26] Yann LeCun. The mnist database of handwritten digits. http://yann. lecun. com/exdb/mnist/, 1998. [27] Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny images. 2009. [28] Tom Fawcett. An introduction to roc analysis. Pattern recognition letters, 27(8):861–874, 2006. [29] Tinghui Ouyang, Vicent Sanz Marco, Yoshinao Isobe, Hideki Asoh, Yutaka Oiwa, and Yoshiki Seo. Corner case data description and detection. In 2021 IEEE/ACM 1st Workshop on AI Engineering-Software Engineering for AI (WAIN), pages 19–26. IEEE, 2021. [30] Stallkamp, Johannes, Marc Schlipsing, Jan Salmen, and Christian Igel. The German traffic sign recognition benchmark: a multi-class classification competition. In The 2011 international joint conference on neural networks, pages 1453-1460. IEEE, 2011. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Advancing OOD Detection: A Deep Learning Approach

Your Guide For Overcoming Procrastination and Getting Down to Work

Effective Anomaly Detection Pipeline for Amazon Reviews: References & Appendix

Assessing VASP Solvency For Cryptoassets: Conclusions, and References

Stories, News & Mentions about KAIZAN

Kaizan’s Stories on HackerNoon