<?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 Keldon Alleyne on Medium]]></title>
        <description><![CDATA[Stories by Keldon Alleyne on Medium]]></description>
        <link>https://medium.com/@keldonalleyne?source=rss-1166c90205a2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*_umE0_0tj0icCCxnSCLuaQ.jpeg</url>
            <title>Stories by Keldon Alleyne on Medium</title>
            <link>https://medium.com/@keldonalleyne?source=rss-1166c90205a2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 04:05:12 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@keldonalleyne/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[Why we need to talk about Ron Schnell’s “charts” in the Johnny Depp v Amber Heard Defamation Trial]]></title>
            <link>https://keldonalleyne.medium.com/why-we-need-to-talk-about-ron-schnells-charts-in-the-johnny-depp-v-amber-heard-defamation-trial-c2e3bdc90608?source=rss-1166c90205a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/c2e3bdc90608</guid>
            <category><![CDATA[johnny-depp]]></category>
            <category><![CDATA[fastai]]></category>
            <category><![CDATA[expert-witness]]></category>
            <category><![CDATA[statistical-analysis]]></category>
            <category><![CDATA[amber-heard]]></category>
            <dc:creator><![CDATA[Keldon Alleyne]]></dc:creator>
            <pubDate>Thu, 19 May 2022 23:01:57 GMT</pubDate>
            <atom:updated>2022-05-19T23:31:54.983Z</atom:updated>
            <content:encoded><![CDATA[<p>On Day 19 of the Johnny Depp and Amber Heard Defamation Trial an “expert” witness in statistical analysis, Ron Schnell, gave his expert testimony.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*yJ8lnkleXxMT56Co.jpg" /></figure><p>I won’t repeat how extraordinary he is as a programmer. A Google search will reveal that he began lecturing in college at the age of 14 and has a series of humbling accomplishments — each of which is probably unachievable in the ordinary person’s lifetime.</p><p>But his statistics suck balls.</p><p>He makes textbook mistakes and does a worse-than terrible job at statistical analysis and reminds us why Correlation Is Not Causation.</p><h3>Correlation Is Not Causation</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*QnMm_Eh5L6y_Vhl_.png" /><figcaption>Correlation between non-commercial space launches and Sociology doctorates awarded in the US</figcaption></figure><p>We’re used to causes correlating with their effects.</p><p>You push the cup, it tips over the counter, it falls, it smashes, and its contents sink into the grooves of your kitchen tiles.</p><p>In this case, Schnell finds a correlation between mentions of Waldman and negative tweets about Amber Heard.</p><p>And here’s the fundamental flaw.</p><p>He’s ignoring everything else — including other news coverage, evidence made public, and media related to the case.</p><p>Is Schnell correct and it is purely by the merits of Waldman’s effortless charisma that people are making negative tweets about Amber Heard?</p><p>Or could they be motivated by something else — like seeing video evidence showing that she has been grossly dishonest in her testimonies?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*T7poh_lvqZTCG2mi.jpg" /><figcaption>Amber Heard’s manipulated photo that she persists to deny are from the same image</figcaption></figure><p>I’ve been manipulating images for around 25 years (for fun) and have proven my proficiency in writing image processing algorithms at Queen Mary University of London.</p><p>I’ll eat a bowl of nail clippings if one of those images is not produced by applying filter(s) to the other every day for a month.</p><p>According to Schnell, seeing convincing evidence that Amber Heard is lying could not possibly be a cause for people having a negative opinion of Amber Heard.</p><p>Amber Heard persists to lie even when presented with irrefutable video or audio evidence in court.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/0*m6w1QxR7avyhL-AB" /></figure><p>People do not like that.</p><p>But according to Schnell, they absolutely could not be factors at all.</p><p>According to Schnell, people could not be making references to Waldman because he has eloquently communicated what they already believed to be true based on video and audio evidence showing Amber Heard was lying.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*4Ev3sSlfoE0m6FDW" /><figcaption>The rest of the clip showing Johnny Depp attacking cupboards</figcaption></figure><p>Initially, videos surfaced showing Depp acting violently towards cupboards.</p><p>I was apprehensive about siding with Heard because I wanted to see the video in context as I know how abusers are able to orchestrate scenes like this.</p><p>The rest of the video confirmed my suspicions. She antagonised him.</p><p>She was not a victim. She does not fear Johnny Depp.</p><p>She berates him for complaining about her “decking” him in his jaw.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*I5e3XwPybs1UAYwy.png" /><figcaption>Amber Heard auditions for a role she will never land …</figcaption></figure><p>But according to Ron Schnell — none of this could possibly have caused people to have a negative opinion about Amber Heard.</p><p>I’m hoping Jeremy Howard or Rachel Thomas from Fast.ai will further elaborate on this as I find it concerning that a man with such a poor grasp of statistical analysis was allowed to present himself as an expert in statistics on the merit of his genius programming.</p><p>I’m a programmer. I know how easy it is to write programs that can perform advanced methods of statistical analysis that a statistician might find difficult (or nigh impossible) to write.</p><p>But that doesn’t make one immune to making rookie errors in statistical analysis.</p><p>What’s worse is that his supposed expert authority may mislead a jury, while the field typically requires sufficient skill to recognize these fatal sins.</p><p>Expert witnesses should be subject to a panel of field experts who can give unbiased guidance or corrections in cases like this when the “expert” clearly has overlooked the most basic and fundamental matters.</p><p>An expert witness can do far more harm in misleading the jury than a series of masterful liars on the stand.</p><h3>Not a Valid Root Cause Analysis</h3><p>There are much more appropriate methods for identifying whether Waldman is the sole cause for people’s negative opinions of Amber Heard.</p><ol><li>Find a group of people not yet exposed to the details of the case.</li><li>Present items relating to the case but in different orders and measures (i.e. some people will never be presented with Waldman’s comments).</li><li>Measure changes in their opinion after being exposed to each item.</li></ol><p>This is how Facebook, Google, YouTube, Twitter, Amazon and friends figure out which adverts, videos and other content to give to you.</p><p>This is also how you can much more accurately measure the impact of a potential cause — all of which seems to Schnell fails to consider or take into account.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c2e3bdc90608" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How the “naughty kid” taught himself to code, microelectronics, and other stuff]]></title>
            <link>https://keldonalleyne.medium.com/how-the-naughty-kid-taught-himself-to-code-microelectronics-and-other-stuff-d92e291105f0?source=rss-1166c90205a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/d92e291105f0</guid>
            <category><![CDATA[diversity-in-tech]]></category>
            <category><![CDATA[self-taught-developer]]></category>
            <category><![CDATA[learn-to-code]]></category>
            <dc:creator><![CDATA[Keldon Alleyne]]></dc:creator>
            <pubDate>Fri, 12 Feb 2021 13:51:52 GMT</pubDate>
            <atom:updated>2023-10-13T17:38:54.311Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Software Developer, Keldon Alleyne, describes his path to teaching himself how to program on an ancient computer before the widespread adoption of the internet and the availability of tutorials.</strong></p><p>The year was ’92, and my mother and I were on our way home from my after-school club where we’d play pool, and box hockey, and every Wednesday we’d be treated to a French lesson from Mrs Donovan.</p><p>Noticing the local thrift store had a “new” second-hand piano I couldn’t resist but tug my mother’s arm to go inside.</p><p>The shopkeeper was impressed with how well I played Slow Rock despite never taking piano lessons, and the following week gave me a bucket of computer magazines after my mother had told him how much I was into video games.</p><figure><img alt="Level map for Sonic The Hedgehog on the Sega Mega Drive" src="https://cdn-images-1.medium.com/max/1024/0*jKCluAZQYGEISKfW.png" /><figcaption>Level map for Sonic the Hedgehog on the Sega Mega Drive</figcaption></figure><p>The magazines had everything from game guides and level maps to electronic hardware ads and snippets of code written in BASIC.</p><p>We then took a cab to my cousin’s as it was raining that evening, and I spent the entire journey deciphering the meaning of the code snippets under the fleeting and cyclic ambience supplied by street lights.</p><p>I continued to study the code when we arrived at my cousin’s rather than watch cartoons or play video games.</p><p>The level maps, code, and advertisements for electronic equipment sparked the imagination of 7-year-old me. There was no reason why I couldn’t design my own level maps. All I needed was a big enough sheet of paper. And the logic behind the code made a lot of sense. They were executed in order. Some commands like GOTO will send you to the numbered line, and I just needed access to the manual to understand the commands I couldn’t figure out.</p><h4>But I was still the “naughty kid”</h4><p>I wasn’t much of a trouble maker but I was vocal when I didn’t agree, had something of a <em>talent</em> for expressing my anger or frustrations in plausibly deniable ways, and could protest with the persistence of a bull and the clarity of a talcum powder laced slap in the face. I wasn’t an easy kid.</p><p>It would take a few more years for me to dig my teeth into electronics where I would learn how to build and design analogue circuits, and later, digital circuits.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/299/0*v1DYc3bzL9LlUqyT" /><figcaption>Usborne Computer Science books for kids</figcaption></figure><p>At the same time, I’d managed to stumble across a few books on programming at the local library but had no computer to try it on.</p><p>In 1998 I got my first break. A discarded council PC found its way to my uncle’s living room. It was a 17 Mhz 80286 with a 20 MB hard drive, 4 MB of RAM running MS-DOS 6.0 and only had four usable programs on it (none of which my uncle knew how to start).</p><p>Each week when I visited I spent an hour or so figuring out how to navigate DOS and eventually discovered Word Perfect.</p><p>He gave me the PC shortly afterwards, and that was when it all began.</p><p>QBasic was the main language I used to learn to program. Although I’d read lots of books on programming, none of the BASIC code was compatible, and I had to learn it all using the information provided in QBasic’s help section.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/642/0*LUngnhHuxv1oS8n-" /><figcaption>QBasic on MS-DOS</figcaption></figure><p>The other programming environment I found was Microsoft Debug. It’s not intended for writing code, but it was all I had.</p><p>I had no manual, book or tutorial on assembler. Instead, I taught myself through sheer trial and error — browsing the memory for new commands, trying them out, and working out what the command did by examining changes in the registers or memory.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/0*wY67UH1d2yhldwnO" /><figcaption>x86 assembly via debug.exe</figcaption></figure><p>Right now there’s a lot of discussion on how to get more minorities into tech. I’d love to help give others the chance I had, but I think there are many misconceptions about what you need to get started (I had a crappy computer that was older than me), and how much children can understand.</p><p>I wasn’t your typical geek. I was a musician, a bit of a class clown, not too bad at sports, pretty social, got in trouble (mostly talking, talking back, and the “intensity” of my protests), and was mostly just curious about the world and how it worked.</p><p>I wasn’t expecting to get a degree or even work as a programmer. I just wanted to make games and had ambitions to be a designer.</p><p>I did excel in maths, however, which opened the most doors for me. Through one program I learned a lot of advanced maths (encryption, matrices, algorithms, statistics, etc.) that I could apply or just try out for myself in code (a few months later on my 286).</p><h4>Luck and Serendipity (a terrible play on Sense and Sensibility)</h4><p>But there was a lot of luck involved. Had I not found that book on MS-DOS interrupts I’d never have known how to access the mouse and video memory. It was through a classmate that I gained access to a C++ compiler, and later, a lucky find in the library(*1)that I was able to learn how to perform 3D rendering in C++ as well as all the maths involved.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/512/0*XWbAAn_K3ic6E4MI" /><figcaption>ProtoCAD 3d for MS-DOS</figcaption></figure><p>I learned 3D modelling through ProtoCAD and used it to build a 3D model of a CD rack for my GCSE Design and Technology project, which I then “enhanced” with graphical effects via NeoPaint (all provided by the same friend who had a drawer of floppy disks his father had collected over the years).</p><p>Seeing my friend playing Pokémon on an emulator he downloaded with his ADSL internet convinced me that there must be a way for me to make my own console games — introducing me to the world of homebrew games development that landed me my first developer role nearly a decade later.</p><p>A lot of hard work was involved — hours upon hours of me trying to figure out how to make things with nothing more than my own wits and my little notepad.</p><p>But for the most part, the largest factor in all of this was a series of happy accidents that are hard to replicate or even match with any diversity programme.</p><p>I was fortunate to not know I was “too young” to learn 3D graphics by myself or to pick up C++ as a 13-year-old. I truly believed we were all on a level playing field, and that if other people could do it, then so could I.</p><p>My hope is more young children will learn to be as naive as I was.</p><p>*1: C++ Real-Time 3D Graphics, Andrew Tyler (Sigma Press 1994)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d92e291105f0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Some videos for understanding Monte Carlo Tree Search]]></title>
            <link>https://keldonalleyne.medium.com/some-videos-for-understanding-monte-carlo-tree-search-196d34047950?source=rss-1166c90205a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/196d34047950</guid>
            <category><![CDATA[monte-carlo-method]]></category>
            <category><![CDATA[markov-chain-monte-carlo]]></category>
            <category><![CDATA[monte-carlo-tree-search]]></category>
            <category><![CDATA[brownian-motion]]></category>
            <dc:creator><![CDATA[Keldon Alleyne]]></dc:creator>
            <pubDate>Mon, 28 Sep 2020 13:22:23 GMT</pubDate>
            <atom:updated>2020-09-28T13:22:23.245Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Roulette wheel with overlayed image of mathematical graphs depicting Monte Carlo Tree Search" src="https://cdn-images-1.medium.com/max/1024/1*UhlX1RI0mypa-6JtPlfBIg.jpeg" /></figure><p>When I want to learn a new concept, I tend to look at multiple sources to see it from a variety of perspectives.</p><p>The videos shared in this article will help you to connect the dots in your own way.</p><p>It’s important that you seek inspiration rather than instruction if you aim to understand these concepts.</p><p>You will learn much better through inspired trial and error — especially if it leads to you self deriving your own algorithms or formulas.</p><h3>Monte Carlo Methods</h3><p>Named after the home of the world’s most exclusive casino, Monte Carlo methods use randomness to solve a wide variety of problems.</p><h4>LeiosOS: What is Monte Carlo</h4><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FAyBNnkYrSWY%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DAyBNnkYrSWY&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FAyBNnkYrSWY%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/9fbd97f141bfb8741158c1531ac32503/href">https://medium.com/media/9fbd97f141bfb8741158c1531ac32503/href</a></iframe><h4>Patrick Boyle: What is the Monte Carlo method? | Monte Carlo Simulation in Finance</h4><p>For a GUI project, I used Monte Carlo methods to make a fairly convincing graph depicting stock prices. I added a bias using the Golden Ratio applied to the last time period, which somehow made it look a lot like a real stock graph.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FluSEdT2HBI4%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DluSEdT2HBI4&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FluSEdT2HBI4%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/6504eeac0d539b45ae23d5ea76c660af/href">https://medium.com/media/6504eeac0d539b45ae23d5ea76c660af/href</a></iframe><h4>Khan Academy: Monte Carlo Simulation to Answer LeBron’s Question</h4><p>For those with an affinity to basketball, I’ve included this last video on Monte Carlo methods</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F-fCVxTTAtFQ%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D-fCVxTTAtFQ&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F-fCVxTTAtFQ%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/f2bb72f60ae1a817d9e46b72edb3d48f/href">https://medium.com/media/f2bb72f60ae1a817d9e46b72edb3d48f/href</a></iframe><h3>Brownian Motion</h3><p>Ever wondered why you keep walking past the same tree when you get lost?</p><p>Brownian motion holds the answer.</p><h4>FuseSchool — Global Education: What Is Brownian Motion?</h4><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F4m5JnJBq2AU%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D4m5JnJBq2AU&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F4m5JnJBq2AU%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/fcc797457b7d6dbd93ae40de527fcbd8/href">https://medium.com/media/fcc797457b7d6dbd93ae40de527fcbd8/href</a></iframe><h3>Markov Chain and Markov Chain Monte Carlo</h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FXV4yj4T4PBQ%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DXV4yj4T4PBQ&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FXV4yj4T4PBQ%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/2c6de8b6647ab406951ef376a32438fe/href">https://medium.com/media/2c6de8b6647ab406951ef376a32438fe/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F3ZmW_7NXVvk%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D3ZmW_7NXVvk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F3ZmW_7NXVvk%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/a53d2ac4a9c7353e4c6ef2496e438643/href">https://medium.com/media/a53d2ac4a9c7353e4c6ef2496e438643/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fh1NOS_wxgGg%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dh1NOS_wxgGg&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fh1NOS_wxgGg%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c6788f90bcabcddfedeb76688dc42dcf/href">https://medium.com/media/c6788f90bcabcddfedeb76688dc42dcf/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F7QX-yVboLhk%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D7QX-yVboLhk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7QX-yVboLhk%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/06dbaebe85bce6f1fe667a4817ba5a7e/href">https://medium.com/media/06dbaebe85bce6f1fe667a4817ba5a7e/href</a></iframe><h3>Bonus Round: Iterative Deepening</h3><p>It would be a crime to not introduce you to iterative deepening.</p><p>In a nutshell, it uses a depth-limited depth-first search and iteratively increases the depth limit.</p><p>Paradoxically this has the same algorithmic complexity as a depth-first search limited to the depth of the shortest path, but without you ever needing to know the shortest length beforehand.</p><p>Grasping this concept might give you more insight into tree search algorithms and how to come up with your own. For example, how about combining iterative deepening with random-walks?</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FY85ECk_H3h4%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DY85ECk_H3h4&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FY85ECk_H3h4%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/e2c030e014ad22a3a8469b142d24aa82/href">https://medium.com/media/e2c030e014ad22a3a8469b142d24aa82/href</a></iframe><h3>Monte Carlo Tree Search</h3><p>No videos. Instead, I ask you. How would you use randomness to solve a tree search problem?</p><p>Try it out yourself.</p><p>For inspiration I’ve defined a few problems you can solve with MCTS:</p><ol><li>Dominoes AI: it’s a simple game that’s easy to code and has few branching points.</li><li>Blackjack AI</li><li>Roulette wheel AI: this does not initially seem appropriate. But think about this one for a minute, and you’ll see it’s actually quite perfect. For extra credit, draw a graph of the odds of each turn and see if you can draw any conclusions for it, or ways to improve on your algorithm.</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=196d34047950" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>