<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>NEAT on Nikhil Kadiyan</title>
    <link>https://nikhilcad.github.io/tags/neat/</link>
    <description>Recent content in NEAT on Nikhil Kadiyan</description>
    <image>
      <title>Nikhil Kadiyan</title>
      <url>https://nikhilcad.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</url>
      <link>https://nikhilcad.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</link>
    </image>
    <generator>Hugo -- 0.124.0</generator>
    <language>en</language>
    <lastBuildDate>Sat, 12 Sep 2026 10:00:00 +0530</lastBuildDate>
    <atom:link href="https://nikhilcad.github.io/tags/neat/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>I made a neural network learn to dodge bullets, from scratch</title>
      <link>https://nikhilcad.github.io/blog/shooter-neuroevolution/</link>
      <pubDate>Sat, 12 Sep 2026 10:00:00 +0530</pubDate>
      <guid>https://nikhilcad.github.io/blog/shooter-neuroevolution/</guid>
      <description>I&amp;rsquo;ve wanted to implement NEAT (NeuroEvolution of Augmenting Topologies) from scratch for a while now. Every tutorial I found online just does pip install neat-python and calls .evolve(), and that felt like cheating myself out of actually understanding it. So I built a small twin-stick space shooter in C++ with raylib, and instead of playing it myself, I let a population of evolving neural networks play it for me.
No PyTorch, no gradient descent, no backprop.</description>
    </item>
  </channel>
</rss>
