<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Shower Thoughts on tl</title><link>https://blog.tklingard.com/tags/shower-thoughts/</link><description>Recent content in Shower Thoughts on tl</description><generator>Hugo</generator><language>en</language><lastBuildDate>Thu, 30 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.tklingard.com/tags/shower-thoughts/index.xml" rel="self" type="application/rss+xml"/><item><title>Thoughts on Meta Learning</title><link>https://blog.tklingard.com/posts/2026-04/thoughts-on-meta-learning/</link><pubDate>Thu, 30 Apr 2026 00:00:00 +0000</pubDate><guid>https://blog.tklingard.com/posts/2026-04/thoughts-on-meta-learning/</guid><description>&lt;p&gt;One of the questions I&amp;rsquo;ve been wondering recently is&lt;/p&gt;
&lt;blockquote&gt;
&lt;p&gt;For World Models*, what should the model weights contain at the start of a rollout?&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;&lt;em&gt;*I&amp;rsquo;m referring to world models in the robotics sense, rather than the pixel-generation game engine sense&lt;/em&gt;&lt;/p&gt;
&lt;p&gt;This may feel obvious: the world model is, by definition, a model of the world our agent finds itself in. It&amp;rsquo;s reasonable therefore to say so it should contain information on gravity; the definition of a kitchen; where to look for ketchup&amp;hellip; But there is a broader question here which in a sense mimics the nature/nurture debate (what instincts and knowledge are babies born with vs what they learn through observation and interaction), alongside a question of utility - nobody wants to buy a self driving car that&amp;rsquo;s not yet learned how to drive!&lt;/p&gt;</description></item></channel></rss>