<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Crypto Doodler on Medium]]></title>
        <description><![CDATA[Stories by Crypto Doodler on Medium]]></description>
        <link>https://medium.com/@cd01?source=rss-8adbfceb4132------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*mRsVvBP5MZl8X84hciryMQ.png</url>
            <title>Stories by Crypto Doodler on Medium</title>
            <link>https://medium.com/@cd01?source=rss-8adbfceb4132------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 07:37:43 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@cd01/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Kelly Criterion]]></title>
            <link>https://medium.com/@cd01/kelly-criterion-c8fe09c6b77d?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/c8fe09c6b77d</guid>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[kelly-criterion]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[counterintuitive]]></category>
            <category><![CDATA[investment]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Mon, 08 May 2023 13:50:47 GMT</pubDate>
            <atom:updated>2023-05-08T13:50:47.786Z</atom:updated>
            <content:encoded><![CDATA[<p>Kelly Criterion is used to maximize the expected value of the logarithm of wealth. This is equivalent to saying maximizing the expected value of wealth in a given set of choices (bets).</p><p>Let’s start with a counter-intuitive example:</p><ol><li>You are given $100 at the beginning.</li><li>You are to roll a die: 1, 3, 5 you win 80% of your bet amount; 2, 4, 6 you lose 50%. You have to bet all of your money each time.</li><li>You have to play 50 times consecutively.</li><li>Result: you probably would end up with $7. Why? Because the median across 100 participants who played this game is $7. The mean is $69.7k though. (I have run this result once).</li></ol><p>You would ask, why wouldn’t you end up with $69.7k? This is because the median is a better estimator in this case. (Refer to results <a href="https://docs.google.com/spreadsheets/d/1JTXPTsypcVeV10ZAjxQf7Fs6-Pyz_OcOYvI6MRdCLrk/edit?usp=sharing">here</a>)</p><p><strong>Median vs Mean, which to use when?</strong></p><p>Imagine there are 10 people in a room, nine of them bankers of Wallstreet and one Bill Gates. The average net worth of the 9 would be US$2M, with Bill Gates in the pool, the average net worth of them would be US$11.4B.</p><p>If you were to guess the average net worth of an individual in that room, what would you guess? Does the mean of US$11,140,000,000 make sense? I doubt that. I would guess the median of US$2,000,000. This is because the net worth of Bill Gates is an outlier to the estimation of the 9 bankers.</p><p>The same goes for the previous example, given a large enough of participants, there would be a lucky someone who rolled 1, 3, or 5 fifty times consecutively and won a horrendous amount of money, but what are the odds it would be you? If you run this many times, you would find the median to be about $7, but the mean would fluctuate greatly. This means that you would end up with $7 most of the time. The distribution of this game is heavily skewed at both ends. The luckiest winner “contributed” a significant amount to the mean, but those gains mean nothing to you.</p><p>The good news is, there is a mathematically guaranteed way to break this game.</p><p><strong>Enter Kelly Criterion…</strong></p><p>Suppose you are allowed to bet a fraction of your money, what would be that amount to guarantee you would end up with more than $100 after 50 games?</p><p>If you bet 37.5% of your wealth at each game, you are likely to end up with $393 at the end of the game. In fact, out of the 50 participants, approximately 39 of them would end up with more than $100. And as always, there will be some unlucky ones who would end up with almost nothing and some would win large.</p><p>f* = p/a — q/b = 0.5/0.5–0.5/0.8 = 0.375</p><p>where f* = fraction of wealth to bet; p = the odds of success; q = odds of failure; a = lost in the negative outcome; b = gain in the positive outcome.</p><p><strong>In relation to investment</strong></p><p>Kelly Criterion seems like a guarantee of financial success. But it relies on a large premise. That is, the odds of success/failure and the gains/losses are known. This assumption itself is almost impossible to calculate in any investment strategy/product.</p><p>However, suppose there are certain investment products that have a designed risk and return characteristic, perhaps there is a way to construct a portfolio that has “defined” parameters p, q, a, and b. Certainly, this is too ideal to be real. Even if such a portfolio existed, p, q, a, and b should be dynamic given the changing market.</p><p>Regardless, Kelly Criterion provides insight into bet sizing that is optimal in a given situation. The job here is to figure out the success rate and the gains/loss when we are facing such investment opportunities. The important thing is all these are estimations, based on individual best judgment. What separates good from great investors is based on how accurately these parameters are being guessed. Undeniably, there will be lucky fools and unlucky geniuses. But given a long enough time and enough sampling, the median of the net wealth should converge according to how well we guesstimate these parameters.</p><p>Also, another crucial point is to never go bust. Say an investment has a 10000x return, but a 0.00001% to go zero, what portion of your wealth you would bet on it? Short answer, not 100%. More on that (the survival part) later.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c8fe09c6b77d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Summary of The Book “Loonshot”]]></title>
            <link>https://medium.com/@cd01/summary-of-the-book-loonshot-816c0a15bf61?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/816c0a15bf61</guid>
            <category><![CDATA[product-development]]></category>
            <category><![CDATA[management]]></category>
            <category><![CDATA[disruptive-innovation]]></category>
            <category><![CDATA[book-review]]></category>
            <category><![CDATA[loonshots]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Mon, 30 Jan 2023 06:21:41 GMT</pubDate>
            <atom:updated>2023-01-30T06:21:41.379Z</atom:updated>
            <content:encoded><![CDATA[<h3>About Loonshot</h3><p>Loonshot by Safi Bahcall is an easy-to-read book. It is full of stories about the emergence and fading of some well-known products and companies. It provides some key insights into the success and failure of companies. It is an essential read for leaders who would want to build a strong product and navigate their organization in the ever-changing market.</p><h3>🚀 The Book in 3 Sentences</h3><ol><li>Loonshot: A neglected project, widely dismissed, it is the champion written off as unhinged.</li><li>Bush-Vail rules: separate the phases (the groups working on loonshots and on franchises) and create dynamic equilibrium (ensure that projects and feedback travel easily between the two groups). Break apart while staying connected.<br>a) Separate the Phases — Separate your artists and soldiers, tailor the tools to the phase;<br>b) Dynamic Equilibrium — Love your artists and soldiers equally, manage the transfer, not the technology</li><li>Being good at loonshots and good at franchises are phases of an organization — whether that organization is a team, a company, or a nation. That’s what the science of emergence tells us.</li></ol><h3>🎨 Impressions</h3><p>Loonshots and franchises require a sweet spot of constant exchange between the soldiers (the personnel that talks to the market and gathers products requirement) and artists (the personnel that develops the product). It is not enough to be successful in the p-type loonshots, but also successful in the s-type loonshots. The former brings out a successful product, the latter makes the public use it.</p><h3>Who Should Read It?</h3><p>Managers that would like to cultivate an innovative team and run a successful startup. It is impossible to have a successful product without knowing what the market really thinks. The artists are the inventors of the product, the soldiers are those who go into the field and understand the needs of the market, come back to the artists and make the product better.</p><h3>☘️ How the Book Changed Me</h3><p>💡 How my life / behaviour / thoughts / ideas have changed as a result of reading the book.</p><ul><li>It is important to understand the market needs, think like a soldier that goes into the field and gathers that information</li><li>Sometimes it is not the product that failed, it is the market that is not ready (timing of the market, for instance as of writing, Meta’s on metaverse is not necessarily a failure yet, it could be that the market is not ready for it). We need to differentiate the false fails.</li><li>Note to self: The magic number of 150 or some other threshold, has been used to monitor terrorism, such as uncovering groups of people coalescing for certain activities. The same idea could apply to be developed into some sort of trading algorithm or to be designed into certain strategies for company management.</li></ul><h3>✍️ My Top 3 Quotes</h3><ul><li>While individuals remain puzzles, man in the aggregate “becomes a mathematical certainty”</li><li>Had the Song emperors appointed and listened to a Chinese Vannevar Bush, the scientific and industrial revolutions might well have taken place five centuries earlier. And we would all be speaking Chinese.</li><li>Phase transitions are triggered when small shifts in system properties — for example, density or temperature — caused the balance between those two forces to change.</li></ul><h3>📒 Summary + Notes</h3><h3>Part One: Engineers of Serendipity</h3><p>Beware of the false fails. Understand the reason users stay or leave. For example, users weren’t leaving because social networks (like Friendster crashed often) were weak business models. They were leaving because of a software glitch. Friendster did not scale its site, it was built for a few thousand users.</p><p>Two types of loonshots:<br>a) Product type. P-type loonshots. People say. “There’s no way that could ever work.” And then it does<br>b) Strategy type. S-type loonshots. People say, “There’s no way that could ever make money.” And then it does.</p><p>Escaping the Moses Trap (it occurs when the company loses the balanced exchange of ideas and feedback between soldiers and artists):<br>a) Separate the three phases:<br>- watch your blind side: nurture both types of loonshots (product and strategy)<br>b) Create a dynamic equilibrium:<br>- appoint, and train, project champions to bridge the divide<br>c) Spread a system mindset:<br>- identify teams with outcome mindsets, and help them adopt system mindsets</p><h3>Part Two: The Science of Sudden Change</h3><p>Phase Transitions — A sudden change of states, the pivotal moment (a change towards success or decline) a company/product faces when it crosses a certain threshold.</p><p>a) The axiom of emergence: While individuals remain puzzles, man in the aggregate “becomes a mathematical certainty”.</p><p>b) The power in the tail: The evidence for groupthink in markets, from tulip mania to the internet bubble, is strong:<br>- Power law: there were 32 times fewer cliques of 40 people than cliques of 10. There were 32 times fewer cliques of 160 than cliques. And so on. The number of cliques decreased with the size of the clique by an unusual power: 2.5<br>- Power law with an exponent of 2.5 is also seen in data from wars in Iraq and Afghanistan</p><p>c) The magic number 150: When the size of a group exceeds a critical threshold, career interests triumph. That’s when teams will begin to dismiss loonshots and only franchise projects. Company leaders need to look out for these signs and indicators.</p><p>d) Raise the magic number. Leaders could use soft equity: nonfinancial stakes in project success, like peer recognition. Shifting rewards more toward projects and less toward promotion. Bring in a chief incentives officer. Encourage looser controls, more experiments, and peer-to-peer problem-solving.</p><h3>Part Three: The Mother of All Loonshots</h3><p>The three conditions for a loonshot nursery:<br>a) Phase separation: separate loonshot and franchise groups<br>b) Dynamic equilibrium: seamless exchange between the two groups<br>c) Critical mass: a loonshot group large enough to ignite</p><p>Case study: Why does the world speak English?</p><ul><li>England established the earliest example of a successful loonshot nursery inside one country. The secret lies within The Royal Society of London, which was created in 1660. It brought together nearly all founders of modern science in England, including Robert Boyle, Robert Hooke, and Isaac Newton.</li><li>The collective success of disruptive inventions fueled the imperialism of the British Empire between the late 16th and early 18th centuries. It was at that time that the technology had gotten good enough to allow them to travel large distances around the world through maritime and continental.</li></ul><p>Loonshots (Today) vs Disruptive (In Hindsight). In hindsight, we can identify what were a few disruptive products that changed the world and the living habits of humankind. The point is what can we do today to identify (with higher chances of success) loonshots that will be the next disruptive product?</p><ul><li>The transistor: Transistors&#39; goal was to improve existing amplifiers and relay switches. It began as a sustaining innovation and was much more expensive than a vacuum tube. Only sold to high-end customers like the military. Today, it has become the basis of modern electronics and is one of the most important industries.</li><li>Online search: Larry Page and Sergey Brin goal was to create a sustaining innovation to improve algorithm for prioritizing internet search results. Today, Google is one of the most valuable companies on earth.</li><li>Walmart: Sam Walton opened stores in rural areas, far from big cities, and started in Arkansas with a population of just 3000. Today, the Walton family is the richest family in the United States of America.</li><li>Ikea: Kamprad was banned from exhibiting at a trade fair and designers were forbidden to work with him (by other furniture companies). He was forced to hire his own designers and source his wood supply from Poland. Successive challenges forced Kamprad to innovate. Today, Ikea is the world’s largest furniture retailer.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=816c0a15bf61" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Summary of The Book “AI.未来”]]></title>
            <link>https://medium.com/@cd01/summary-of-the-book-ai-%E6%9C%AA%E6%9D%A5-39ac2aa67a86?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/39ac2aa67a86</guid>
            <category><![CDATA[future-technology]]></category>
            <category><![CDATA[venture-capital]]></category>
            <category><![CDATA[book-review]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Wed, 02 Nov 2022 06:22:24 GMT</pubDate>
            <atom:updated>2022-11-02T06:22:24.585Z</atom:updated>
            <content:encoded><![CDATA[<p>This is my attempt to summarize what I have read and find interesting. Note that the book is originally written in Chinese and what I have written here is the interpretation of what I understand. There would be a loss of meanings and misinterpretations, and I am happy to take in opinions.</p><p>AI.未来, is a book written by Dr Kai-Fu Lee. He is the Chairman and CEO of Sinovation Ventures. He was selected as one of the 100 most influential people in the world by Time Magazine in 2013. Dr. Lee has a PhD in artificial intelligence from Carnegie Mellon University and was the president of Google China. He has vast experience in this field in both the US and China.</p><p>🚀<strong>The Book in 3 Sentences</strong></p><ol><li>The 4 waves of AI: Internet AI, Business AI, Perception AI, and Autonomous AI, are the direction we are heading towards. The boundary between the virtual world and reality is going to be blurred.</li><li>The advances of AI are inevitable. Without intervention, the wealth gap and polarity in society are going to widen. We need to reconstruct the social contract of the meaning of work and how we coexist with AI.</li><li>This book serves as a blueprint for the analysis of how humans and AI can coexist and how AI is being developed and understood.</li></ol><h3>🎨 Impressions</h3><p>The book has summarized the major events of AI development and how the cutthroat environment in China has fostered the perfect conditions for the AI industry. Despite China being 3 decades behind when the internet revolution started, its sheer number of researchers since last decade has been closing the gap. There is still much room for development in the field and the important thing is how can we keep ourselves competitive in this happening.</p><h3>Who Should Read It?</h3><p>People who are interested in understanding the development of AI and how did China catch up in this field so quickly. It is crude to jump to the judgment that China Tech companies are only good at imitating the success of Silicon Valley companies. The survivors are champions in understanding the Chinese market and know how to customize their products in catering to the market. In this book, we get a glimpse of how these companies’ services and what was the policy at that time created many golden opportunities in the market.</p><h3>☘️ How the Book Changed Me</h3><ul><li>I was under the impression that China Tech Companies are merely copying the success of silicon valley products.</li><li>The strategies of tech companies from both worlds are so different. In silicon valley, companies emphasize developing a single superior product to cater to all users. It is common to build a platform and let other companies use their platform to solve real-world practical problems. In China, companies focus on highly customized products that cater to local needs. As a result, iterations happen quicker and competitions are stiffer. They focus on the integration of solutions from the up to downstream.</li></ul><h3>✍️ My Top 3 Quotes</h3><p>**I have written these down in Chinese as I think it is best understood or enjoyed (at least to me) in their raw form.</p><ul><li>人工智能是“演算的巨人，行动的矮子”</li><li>星云法师对开复说到：“人类不能一刻不停的计算、量化一切事物，会侵蚀掉我们内心本真的东西，会阻碍生命中真正的能量的涌现 — 那就是爱。”</li><li>我们应该将公共政策与个人价值融合在一起，花点儿时间研究如何重新定义衡量方法，比如不丹所追求的“国民幸福指数”（Gross National Happiness）。</li></ul><h3>📒 Summary + Notes</h3><h3>Part One: 中国的斯普特尼克时刻 (China’s Sputnik Moment)</h3><ul><li>The Sputnik crisis was a period of public fear and anxiety in Western nations about the perceived technological gap between the United States and the Soviet Union caused by the Soviet launch of Sputnik 1, the world’s first artificial satellite.</li><li>Machine Learning — using input data and training the model to differentiate the relationship between the expected output (cat vs not cat, won a match or lost a match).</li><li>Deep learning is “narrow AI”, it only can be used in a specific field to make decisions, predict and classify. This is different from general AI.</li><li>Deep learning most suitably can be used in the insurance, and loan industry.</li><li>We mistakenly think that the US has the advantage in the AI field. This is because we still think that we are in the “invention era”. But instead, it is the same model being applied in different fields, such as using reinforcement learning and transfer learning in new fields.</li><li>Only if the research can be adopted in real-world applications, then it has value. Practical application is what China is strong at.</li><li>China’s advantage: it is particularly good at adapting products to local niche markets, this is different from the silicon valley idea of making ONE good product for all markets. Online To Offline, O2O innovative business model is a good example of the super app WeChat. In the US, they would prefer to have different apps for different usages.</li><li>The crises in AI: AI industry is highly included towards winners take all. This is exacerbating the problem of economic inequality and unemployment. This is caused by the positive feedback loop achieved by the giants and this would enlarge the gap between them and the smaller companies. More data → better AI → more users → better profit → more models and AI → more data.</li><li>AI New World Order: polarity of technology between the US and China. They would acquire all the global wealth and consumers. In addition, more automation in the country would disrupt the advantages that emerging countries have — cheap labor.</li></ul><h3>Part Two: 从竞技场杀出的世界级创业者 (The survivors of the world-class entrepreneurs)</h3><ul><li>“The Cloner” 王兴 (Wang Xing) brought Facebook and Groupon back to China into 人人网 (Renren) and 美团 (Meituan). This is opposed to the innovative spirit of silicon valley. But the successes of both clones are due to the fact that they are highly customized according to local culture. According to Wang, imitation is a piece of the puzzle in innovation as well.</li><li>In order to compete in China’s cutthroat market, China entrepreneur has to iterate their product much faster and participate in even “unreasonable” price war. The Chinese workers are working at unbelievable hours and speed. However, this results in them putting emphasis on turning research results into practical products.</li><li>The success of these “cloners” is not a result of mere imitation, but they are the distillation and champion of this intense competition.</li><li>The Chinese culture of memorized material and imitation has a long historical influence. In the past, the public examination required candidates to memorize word by word of classic literature. Confucious thoughts recommend pupils to follow the noble and teachers. Moreover, the “scarcity mentality” in the 20th century has motivated the Chinese to work hard and innovate. The one-child policy has resulted in a child receiving the resources from 6 parents and also bearing the expectation of these parents.</li><li>Deng Xiao Pin invoking “to let some people get rich first” has resulted in the phenomenon of the Chinese seeing some of their peers get rich quickly. This has caused anxiety among them to grab whatever opportunity they have.</li><li>in 1998, only 0.2% of Chinese had access to the Internet (in the US 30%). It did not have the condition in the US (Bill Gates founded Microsoft in 1975). Chinese innovators can reference solutions overseas and learn and imitate. But they are not merely imitating, they need to iterate product way quicker than in the US to win the market. They understand what the Chinese like and do not like, and optimize products for the mass.</li><li>For example, Chinese users have a strong personal preference, they are not used to those products from silicon valley that has a “global” standardized product.</li><li>Alibaba vs eBay: Jack Ma created Alipay to solve the confidence issue buyers have — only releasing payment after confirmation of goods. And Alibaba’s free + value-added biz model has won the Chinese market.</li><li>How Baidu and Google open up new tabs (Baidu opens the clicked link in a new tab) is also an example of the difference between browser users of the Chinese and the West. This is designed based on the behavior of the browser users.</li></ul><h3>Part Three: 中国的另类互联网世界 (China’s Information Age)</h3><ul><li>Data: The data Chinese company can collect and are allowed to collect is more than America and Europe combined. This has been critical for applications for AI companies.</li><li>WeChat: the invention that was dismissed by a lot of silicon valley companies. Its innovation includes being the first messaging app for smartphones in China, video call, and voice call early in 2011–1013 (WhatsApp only has it in 2016). Its in-app app is also a big hit. As a result, many service companies do not develop their own app and rely on developing within the WeChat app. Payment offline and online is also an innovation.</li><li>O2O in silicon valley and in China: “going light” or “going heavy”. In silicon valley, it is preferred to go light as they believe the Internet is mainly for information sharing and reducing the knowledge gap. They would build platforms and let other companies deal with real-world problems (via solutions operating on their platforms). In China, it is inclined toward going heavy. They not only build a platform but also include seller, processing, delivery, provide maintenance service, and control payment activity. If needed, they would even subsidize the process to acquire a user base quickly. As a result, they are very detail oriented (which comes with an extraordinary cost). Going heavy also means building a larger moat. A good example is Yelp vs 大众点评 (DianPing). Both were founded in 2004, but Yelp still focuses on reviewing restaurants as their main business. But the latter has included group purchasing, payment, developing supplier relationships, and launching a lot of subsidizing activities.</li></ul><h3>Part Four: 两国演义和七巨头 (The Two Worlds and Seven Giants)</h3><ul><li>7 giants: Google, Facebook, Amazon, Microsoft, Baidu, Alibaba, and Tencent.</li><li>The general perception of AI research is that they are willing to open and share their algo, data, and results. It is also different from many physical sciences, as AI results are completely repeatable and compared in all “labs”.</li><li>Chinese AI researcher has been cited more and more (from 23.2% in 2006 to 42.8% in 2015).</li><li>The 7 giants are playing the role of government in the AI world. They devoted massive resources to developing systems, talent, and resource control, therefore also emphasizing on keeping these results in-house.</li><li>China’s local government (encouraged by the party) has also actively invested in setting up special zones and incubators. For example, Nanjing city came up with policies to attract AI companies and talent to the city. Its setup of “AI Valley” is the epitome of China’s focus on AI development.</li><li>Autonomous driving dilemma: although ADAS can save countless lives by reducing accidents, it still could not answer the ethical problem. For example, between crashing into 2 people with 55% and crashing one person with a 100% chance, which one should it choose? When an accident is bound to occur, should the AI prioritize the driver or the pedestrians? Every risk and scenario involves delicate ethical problems. All these would bring forward long tedious lawsuits, and thus hinder the process of enacting laws for ADAS.</li><li>The US has been good at being open to immigrants and treasuring talents. Hence, it is able to attract talents from all over the world (for example Enrico Fermi and Albert Einstein). China, throughout its rapid economic development, has used a top-down approach to propel development forward. AI being a unique tool and solution has been able to create a new social order and economic system that is advantageous to China.</li></ul><h3>Part Five: 人工智能发展的四波浪潮 (The Four Waves of AI)</h3><ul><li>iFLYTEK. A company that is extremely good at speech recognition, speech synthesis, image recognition, and machine translation has beaten teams from Google, Deepmind, Facebook, and IBM in the “第二语言” competition.</li><li>4 waves of AI: Internet AI, business AI, perception AI, and autonomous AI. Every wave is going to disrupt many different industries and result in AI being more entwined into our daily life. The first two waves are already around us. Internet companies are using algo to replace actuaries, stock brokers, and medical assistants.</li><li>Now, perception AI is learning our faces, recognizing objects around us, and understanding our needs. It is going to blur the boundaries between real and virtual.</li><li>Internet AI: started 15 years ago, and became mainstream in 2012. An example is search engines. The algo starts to understand, study and learn our personal preference and then recommend contents to us. We start to think that the Internet is getting better.</li><li>Business AI: Labeling data. Biz AI labels data that companies have collected for decades and uses them for analysis. An example is insurance companies detecting frauds, banks using it to run credit scores, medical data to improve diagnostics, etc. This wave of AI is better than humans at detecting weak features. Humans are only good at detecting strong features to see the casual relationships between observations. Oftentimes, humans cannot even explain these weak relationships (eg. a debtor that obtains loans on a Wednesday defaults less), but the features have been helping humans make better decisions.</li><li>Perception AI: To machine, data are just 1 and 0. Perception AI changes that. Algo now can imitate the human brain in making sense of data and making decisions. This wave is to extend AI into our daily lives by using immense sensors and AI instruments. An example is Xiaomi smart speaker and Amazon Echo can digitize environment sound, Alibaba “城市大脑 (City Brain)” can digitize transportation data through cameras and object classification, Face++ can digitize facial features and use that to protect virtual wallet.</li><li>This will continue to encourage the environment of Online-Merge-Offline. In this process, every boundary between the digital world and the real would be going to be bridged, by bringing online convenience to offline and bringing offline perception to online.</li><li>Autonomous AI: it is the synthesizer and pinnacle of the previous 3 waves. An example is Autonomous Driving. But the implication is much further. Autonomous AI will completely change how we live, but it will come gradually. It can only first appear in structured places where the value creation is highly valued, eg factories, warehouses, and farms. This is still far from the automation we have today, which is only capable of highly repetitive tasks. Autonomous AI happens when we equip AI with machine vision, and motion sensors and optimize the data collected, then the tasks machine can handle a much wider scope of work.</li><li>Internet AI leading companies: Google, Amazon, NetEase, Baidu, Facebook, Taobao, WeChat, JD, meituan</li><li>Business AI leading companies: Palantir, IBM Watson, Element AI, 4 paradigm, AInnovation</li><li>Perception AI leading companies: Amazon Echo, Face++, Megvii, Amazongo, 小鱼在家</li><li>Autonomous AI leading companies: Tesla, Uber, Waymo, Moment, Jingchi.ai, UISEE</li></ul><h3>Part Six：乌托邦、反乌托邦和真正的人工智能危机 (The Utopian vs Dystopian and AI Crises)</h3><ul><li>We might think that AGI (Artificial General Intelligence) is within our reach. This resulted in two camps in the field — the utopian and the dystopian.</li><li>The utopians see this singularity for humankind as a huge step towards achieving eternity. Some also argue that this could help us unravel the mystery of the universe, solve global warming and cure cancers. It will become the almighty God itself.</li><li>The dystopian would argue this is just the beginning of creating the demon and scenario like The Terminator.</li><li>Views toward unemployment and inequality: the optimist would quote Luddite Fallacy (in 19th century England, the industrial revolution disrupted many jobs and was protested by workers, but the revolution move forward and brought prosperity to their offspring in the coming 2 centuries).</li><li>GPTs (General Purpose Technologies) — are viewed as truly significant technologies. Only 3 inventions are being widely supported as GPT: steam engine, electricity, and Information and Communications Technology (ICT).</li><li>Steam engines and electricity initiated the first and second industrial revolutions. ICT has caused the most benefits to be reaped by the top 1%. But ICT is different from the former 2, it is skill bias and broke the information barrier and helped the world’s top researcher to consolidate knowledge.</li><li>Winner vs Loser: Jobs that require creativity/decision-making and high social ability are the hardest to be replaced by AI.</li><li>Moravec’s Paradox: AI is better at imitating adults’ knowledge and computation power, but is extremely bad at initiating babies’ perception and raw senses.</li><li>AI is a threat to the white collars as the algos are faster and more accurate than humans. However, robots are harder problems as they require mechanical engineering, perception AI, and accurate feedback control. Even when such machines are invented, it still needs to be tested, installed, calibrated, and maintain at sites. Algos can be corrected remotely, but all machinery faults need to be fixed on-site.</li></ul><h3>Part Seven：一个癌症患者的思考 (Reflection of a Cancer Patient)</h3><ul><li>It is believed that the future of AI is independent thinking, complimented by love. Only if they can synergies, then can we preserve the basis of humanity and use AI to create a better world</li><li>We need to rethink and construct what we want as a future and work towards a harmonious and more humane society.</li></ul><h3>Part Eight：人类与人工智能共存的蓝图 (How Humans Coexist with AI)</h3><ul><li>A new social contract: it is predicted that in 15 years 40% — 50% is replaceable. But “technically replaceable” does not mean that they will be replaced overnight. But the general trend predicted is that unemployment would rise and the wealth gap would widen.</li><li>This polarity would cause instability in society. To avoid crises, we need to use wealth created by AI to reconstruct the economic system and rewrite the social contract. A cultural paradigm shift is needed in defining what is the meaning of life as understood by most (like a sense of achievement in work). Without this, the link between humans and society would break. We need to get rid of the notion that work is equal to life. We should emphasize love between humans, empathy, and the willingness to serve others.</li><li>3R: Retraining workers, reducing work hours, and redistributing income. These are used to regulate the parameters of the job market such as skills, time, and reward.</li><li>Universal Basic Income (UBI) and Guaranteed Minimum Income (GMI). Some think that UBI is the solution to the polarity problems caused by these tech companies. However, the issues with UBI do not lie within advocators, but within itself. For instance, a simple UBI rewards people who did not contribute and would discourage work and innovation. Moreover, even though people who receive UBI use that income to retrain themselves, it is not guaranteed that the new skills they acquire would not be replaced by another AI.</li><li>We need to utilize AI’s advantages to create a social system that encourages AI’s development and realize what distinguishes us from machines, i.e the capability to love.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=39ac2aa67a86" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The emergence of the Decentralized Reserve Currency (DRCP)]]></title>
            <link>https://medium.com/@cd01/the-emergence-of-the-decentralized-reserve-currency-drcp-a7b9d62cd377?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/a7b9d62cd377</guid>
            <category><![CDATA[olympus-dao]]></category>
            <category><![CDATA[staking-rewards]]></category>
            <category><![CDATA[decentralized-finance]]></category>
            <category><![CDATA[game-theory]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Wed, 30 Mar 2022 14:02:40 GMT</pubDate>
            <atom:updated>2022-03-30T14:02:40.364Z</atom:updated>
            <content:encoded><![CDATA[<p>The vision of many Decentralized Reserve Currency Protocol (DRCP) is to build a policy-controlled currency system that moves away from the centralized governing system. Normally, a Decentralized Autonomous Organization (DAO) is deployed to manage the currency policy. This means that the policy is decided by the holders of the currency rather than a handful of elites (The Central Banks). This is a key difference.</p><p>“Reserve Currency” refers to a currency that is essentially backed by a reserve, like how the Central Banks use Gold Standard to back their currencies. DRCP backs their currency with a pool of cryptocurrency assets. This gives the holders a further layer of assurance as the token would not fall below the floor value of the backing assets.</p><p>A currency must also fulfill a few key <a href="https://www.toppr.com/guides/business-economics-cs/money-and-banking/qualities-of-good-money/">characteristics:</a> General acceptability, Portability, Durability, Divisibility, Homogeneity, Cognizability, and Stability. A decentralized reserve currency fulfills all of the above except General acceptability and Stability.</p><p>A DRCP must be built to address the characteristics mentioned above. Given the intangibility of crypto assets and the lack of central authorities to back them, the protocols must come up with a creative way to attract holders to accept the currency/token as a store of value. In addition, it has to be sustainable economically. As such, DRCP works on a <a href="https://medium.com/coinmonks/what-is-protocol-owned-liquidity-a-primer-on-the-model-developed-by-olympus-dao-55368f200d66">Protocol Owned Liquidity</a> structure, where the fees to provide liquidity are earned for the Protocol, acting as another source of revenue for the Treasury. Under this schema, higher incentives are in place to attract users and liquidity providers. By amassing a large enough user base and liquidity pool, BOOM, you have the formula for general acceptability and stability right there.</p><p><strong>A grand behavioral economics experiment</strong></p><p>OlympusDAO (OHM) is the first DRCP that offers lucrative high yields to attract users. Early adopters are rewarded with APY up to 8000% by staking their OHM tokens. You probably heard about (3,3). It is mostly a marketing gimmick that stems from game theory. If everyone plays the game co-operatively, i.e. stakers do not sell and the bonders keep increasing the treasury, everyone wins. The bonders win from either the rising or static price movement as the protocols allow them to purchase the native tokens at a discount. The stakers benefits from the APY — they are rewarded with more native tokens of the protocol. OlympusDAO documentation states that if both actors act in a beneficial (rational?) way, then the actor causing a price movement also gets half of the benefit. If both actions are contradictory, the bad actor receives half of the benefit while the good actor gets half of the downside. If both actors sell both get half the downside. So game theory dictates that participants are inclined to stake their positions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3kIIB3kpEed9rSnE" /><figcaption><em>The Payoff Matrix. Source: OlympusDAO</em></figcaption></figure><p><strong>What can DRCP solve?</strong></p><p>DRCPs are a demonstration of DeFi community in building a new financial system. Participants work together towards the vision of creating a stable currency. The power of a currency comes from its ability to be accepted by others for the exchange of goods and services. The ultimate goal is that decentralized reserve currency can one day be used by anyone to purchase a cup of coffee or accepted by a barber in exchange for a haircut, like how a fiat currency is being used. Being deployed on the Blockchain adds a dimension that it is a far more transparent system than any existing centralized financial system. The beauty of DAO lies in its democratic way of deciding the future of the protocols. The future of DERPs is entirely in the hands of their stakeholders.</p><p><strong>What are the risks?</strong></p><p>“What happens in Blockchain stays in Blockchain”. Being decentralized also means that the participants of the protocol are likely to remain anonymous, including the dev team. Nor the protocol needs to comply with any laws or policies of any jurisdiction. Both the upsides and downsides are born by the token holders themselves. There are no enforcers to protect the investors if there are any “RugPulls” occurred, i.e. a project being abandoned and no one is accountable for the lost investments. Another major problem with the anonymity of DeFi projects is that it is a mystery if these protocols are designed by qualified individuals. Therefore, there is no assurance that these protocols are sustainable or make economical sense. Yes, yes, all financial instruments suffer from this problem, despite being endorsed by hyper-qualified individuals. A notable fiasco will be the default of the <a href="https://www.investopedia.com/terms/l/longtermcapital.asp#:~:text=Long%2DTerm%20Capital%20Management%20(LTCM)%20Demise,that%20it%20hold%20its%20positions.">Long-Term Capital Management</a>. Therefore, it is important to DYOR.</p><p><strong>Summary</strong></p><p>Game theory plays a rather significant role in the underlying success of DRCP, especially at its embryonic stage. The game theory aspect is something investors have to factor in when deciding to take on these projects. One thing is certain is that the transparency of the treasury backing mechanism and projects that drive towards decentralization are a step up to a more profitable bank-less separation. It is still too early to say if these protocols are here to stay, but they have attracted a large base of an enthusiastic community, which is the key to driving DeFi forward.</p><p>Thanks for reading. Cheers.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a7b9d62cd377" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Week 2 — More Buzzwords]]></title>
            <link>https://medium.com/@cd01/week-2-more-buzzwords-1bd4d4f67e95?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/1bd4d4f67e95</guid>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[liquidity]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Mon, 21 Mar 2022 15:00:02 GMT</pubDate>
            <atom:updated>2022-03-21T15:03:03.868Z</atom:updated>
            <content:encoded><![CDATA[<h3>Week 2 — More Buzzwords</h3><p>Following our week 1 article, we should expand our list of buzzwords on cryptocurrencies.</p><p>This week, we explore what Fungible Token, NFT, POW, POS, Gas Fee, Market Cap/FDMC, and Liquidity are.</p><p><strong>Fungible Token</strong></p><p>Fungible is a synonym for interconvertible. A fungible token is just like fiat money in its sense of fungibility. This means they can be traded or exchanged for one another. They are divisible into smaller units and are equal in value — one dollar is always worth another dollar; 0.1 Bitcoin is always equal to another 0.1 Bitcoin. Crypto’s fungibility makes it a trusted means of conducting transactions on the blockchain.</p><p><strong>Non-Fungible Token (NFT)</strong></p><p>Artwork is usually deemed non-fungible as each artwork is supposedly one-of-a-kind, such as there is only one Mona Lisa. NFT works similarly as each NFT represents a unique digital asset. There are no two NFTs that are the same; hence, NFTs are non-interchangeable. This is in contrast to crypto-assets like Bitcoin which are fungible/interchangeable in nature.</p><p><strong>Proof of Work (POW)</strong></p><p>POW is a consensus model in blockchain that ensures the validity of a transaction by solving complex cryptographic mathematical equations using computer power. Older cryptocurrencies such as Bitcoin and Ethereum use POW to validate new transactions. Under POW, miners/validators must do “work” that is difficult for them to contribute, but easy for the broader network to verify. The complex equations ensure that no individual miner is able to cheat the system; hence, the network is secured under the collective work of all miners.</p><p><strong>Proof of Stake (POS)</strong></p><p>POS is another method to confirm transactions on the blockchain. In POS, validators provide their digital tokens upfront to earn the right to validate new transactions on the blockchain. The miners are selected based on how many units they provided upfront (their stake). The stakes each miner contributed ensure that no miner would cheat the system as their stake would be “confiscated” if found cheating.</p><p><strong>Gas Fee</strong></p><p>Gas Fee is calculated based on the computational effort required to execute specific operations on any blockchain network. These fees provide incentives for people to mine crypto, as the fees are typically paid to miners who validate transactions. It is a mechanism to secure the blockchain network. By requiring a fee for every computation executed on the network, bad actors are deterred from spamming the network.</p><p><strong>Market Capitalization (Market Cap) / Fully Diluted Market Cap (FDMC)</strong></p><p>In the TradFi context, the market cap is the total market value of a publicly-traded company’s outstanding shares.</p><p>Market Cap = stock price × total number of shares outstanding</p><p>In the DeFi (for the definition of TradFi &amp; DeFi, you can refer <a href="https://medium.com/@cd01/week-1-crypto-buzzwords-b6051ec9655d">here</a>) context, the market cap is the total value of all the current tokens in circulation.</p><p>Market cap = token price × circulating supply</p><p>As of current writing, the largest cryptocurrency by market cap — Bitcoin has a market cap of over US$ 777B. Most DeFi projects do not release all their tokens into circulation at launch. Normally, these projects follow an emission schedule while the remainder is still “locked up” under reserve. This brings us to the definition of Fully Diluted Market Cap (FDMC).</p><p>FDMC is the total value of the maximum supply of tokens.</p><p>FDMC = token price × maximum supply of tokes</p><p>Take Bitcoin as an example, the maximum supply is fixed at 21M and currently has 18.9M in circulation. Therefore, the FDMC of Bitcoin as of current writing is US$ 850B.</p><p><strong>Liquidity</strong></p><p>According to Wikipedia, liquidity is a market’s feature whereby an individual or firm can quickly purchase or sell an asset without causing a drastic change in the asset’s price. Cash is the most liquid asset because it can be exchanged for goods and services instantly at face value. On the other end of the spectrum, we have the painting Salvator Mundi that you just bought. Chances of you securing a fair price and an immediate buyer are reasonably low. This means that the painting is illiquid. Liquidity is often a concern for many cryptocurrencies due to their small market cap and little user bases. For a cryptocurrency to serve its purpose as a currency, it must be liquid enough for users to exchange for goods and services quickly and at face value.</p><p>With these in mind, we will dive deep into the importance of liquidity in the world of DeFi next week. Thanks for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1bd4d4f67e95" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Week 1 — Crypto  Buzzwords]]></title>
            <link>https://medium.com/@cd01/week-1-crypto-buzzwords-b6051ec9655d?source=rss-8adbfceb4132------2</link>
            <guid isPermaLink="false">https://medium.com/p/b6051ec9655d</guid>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[defi]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[web3]]></category>
            <category><![CDATA[decentralized-exchange]]></category>
            <dc:creator><![CDATA[Crypto Doodler]]></dc:creator>
            <pubDate>Tue, 15 Mar 2022 09:22:42 GMT</pubDate>
            <atom:updated>2022-03-15T09:22:42.562Z</atom:updated>
            <content:encoded><![CDATA[<h3>Week 1 — Crypto Buzzwords</h3><p><strong>Prologue</strong></p><p>This is a space to consolidate my research on cryptocurrencies and any topics (mostly cryptos) that I find interesting. I have two goals. First, this serves as a formal log for my research and an audit of my research materials. Second, it is my hope that I can reach out to people that share my passion in this regard and we could learn more together.</p><p>This is the first article. First, we need to lay some groundwork on what cryptocurrencies are about. To understand a thing we must first understand its name. We have heard many things about cryptocurrencies. We have heard a lot of words. But what do they mean? This article is here to explain them as simply as possible. Today we will talk about Cryptocurrency, Web3, TradFi, DeFi, CEX, DEX, Smart Contract, Stablecoin, and Altcoin.</p><p><strong>Cryptocurrency</strong></p><p>Cryptocurrency (crypto) is any form of currency that works as a medium of exchange through a computer network and uses cryptography to secure transactions. It does not rely on a central authority to verify the transactions nor rely on a regulating authority to issue them. Some of the best-known cryptos include Bitcoin — the first cryptocurrency and Ethereum — a blockchain platform that leverages smart contracts for creating decentralized applications. Their market cap totaled to be slightly over 1 trillion dollars as of current writing.</p><p><strong>Web3</strong></p><p>Web3 refers to a decentralized online ecosystem based on blockchain. Its decentralization nature allows users to participate in the network without compromising their personal data. Web3 is the collective ownership of the network by the users. Web1 is characterized by its static and read-only nature. Web2 has evolved to be interactive in which users can read from and write on the network. Web3 is about verifiability, in which users can read from, write on, and own (hence trust) the network. Web3 inherent the characteristic of decentralization and transparent characteristics through the usage of blockchain.</p><p><strong>TradFi</strong></p><p>Traditional finance. They are our familiar retail, commercial, and investment banks. It is the finance world we are familiar with. Think of examples like central banks and FinTechs companies. TradFi is centralized and is governed by well-established rules and regulations.</p><p><strong>DeFi</strong></p><p>Decentralized finance. As opposed to TradFi is governed by centralized authority, DeFi is governed by computer algorithms. DeFi lacks intermediate bodies and carries out transactions through the implementation of smart contracts.</p><p><strong>CEX</strong></p><p>Centralized exchange. They are the third parties that act as brokers between buyers and sellers of cryptos and operate under an order book model. They have more stringent KYC standards and generally offer supplementary services like crypto asset custody. Some of the best-known CEXs are Binance, FTX, Huobi, Coinbase, etc.</p><p><strong>DEX</strong></p><p>Decentralized exchange. DEX allows users to execute peer-to-peer (P2P) transactions without the need for an intermediary. It operates under an automated market maker (AMM) model in which transactions are handled by smart contracts. Asset prices in DEX are determined by these smart contracts and liquidity pools rather than buyers and sellers. Some of the top DEXs are Uniswap, SushiSwap, Raydium, etc.</p><p><strong>Smart Contract</strong></p><p>Smart contracts are programs stored on blockchains that are self-executing once the predefined conditions are satisfied. It can be programmed to execute trades of any kind of crypto assets and remove all intermediaries from the process. All trades are trackable, transparent, and immutable. It is widely used to automate workflow and trigger the next action when predetermined conditions are met.</p><p><strong>Stablecoin</strong></p><p>A class of cryptocurrency that has its price pegged to another class of asset, such as gold, fiat money, etc. The most common stable coins are USDT and USDC, they are backed by dollar-denominated assets of at least equal fair value to the respective coins in circulation.</p><p>This means that you can always redeem 1 USD for each USDT or USDC from the issuer. This also means stablecoins are free from volatility, unlike other cryptocurrencies.</p><p><strong>Altcoin</strong></p><p>Altcoin is short for alternative coins. It refers to other cryptocurrencies other than Bitcoin. There are many types of altcoins. Stablecoins mentioned are one of them. Other examples include Governance Token (allow owners to vote on decisions that influence the specific blockchain ecosystem, eg. Luna), Utility Tokens (tokens that serve some use within a specific ecosystem, eg. Filecoin), Security Tokens (a record of ownership of a real asset with value), etc.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b6051ec9655d" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>