<?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>Continual-Learning on Sanketh's Blog</title><link>https://sankethbk.github.io/blog/tags/continual-learning/</link><description>Recent content in Continual-Learning on Sanketh's Blog</description><generator>Hugo -- 0.169.0</generator><language>en-us</language><lastBuildDate>Mon, 19 Oct 2026 00:00:00 +0530</lastBuildDate><atom:link href="https://sankethbk.github.io/blog/tags/continual-learning/index.xml" rel="self" type="application/rss+xml"/><item><title>Lab 1.2: Teach It 1s and 2s First, Then Introduce 3 — Watching Weights Reorganize</title><link>https://sankethbk.github.io/blog/posts/ml/labs/2026-10-19-lab-2-continual-learning-1-2-then-3/</link><pubDate>Mon, 19 Oct 2026 00:00:00 +0530</pubDate><guid>https://sankethbk.github.io/blog/posts/ml/labs/2026-10-19-lab-2-continual-learning-1-2-then-3/</guid><description>&lt;h1 id="lab-12-teach-it-1s-and-2s-first-then-introduce-3--watching-weights-reorganize"&gt;Lab 1.2: Teach It 1s and 2s First, Then Introduce 3 — Watching Weights Reorganize&lt;/h1&gt;
&lt;p&gt;&lt;strong&gt;MNIST internals series.&lt;/strong&gt; In &lt;a href="https://sankethbk.github.io/blog/posts/ml/labs/2026-10-19-lab-1-mnist-from-pixels-to-neural-network/"&gt;Lab 1.1&lt;/a&gt; we established what a weight row &lt;em&gt;is&lt;/em&gt;: a learned photograph of a class, contrast-carved against its rivals, painted by the update rule
$W_c \leftarrow W_c - \eta\,\tfrac1N\sum (\hat p_c - y_c)\,x^\top$
.&lt;/p&gt;
&lt;p&gt;Now the experiment that motivated the series: &lt;strong&gt;start the model in an impoverished world — only 1s and 2s exist — then let it discover 3.&lt;/strong&gt; This is &lt;em&gt;continual learning&lt;/em&gt; in miniature, and it lets us watch, directly and quantitatively:&lt;/p&gt;</description></item></channel></rss>