<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Model Interpretability on Programmer.ie: Modern AI programming</title>
    <link>http://programmer.ie/tags/model-interpretability/</link>
    <description>Recent content in Model Interpretability on Programmer.ie: Modern AI programming</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Fri, 17 Apr 2026 14:27:45 +0100</lastBuildDate>
    <atom:link href="http://programmer.ie/tags/model-interpretability/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The Eye That Sees</title>
      <link>http://programmer.ie/post/eye/</link>
      <pubDate>Fri, 17 Apr 2026 14:27:45 +0100</pubDate>
      <guid>http://programmer.ie/post/eye/</guid>
      <description>&lt;blockquote&gt;&#xA;&lt;p&gt;&lt;em&gt;Using AI to Decode Symbols Without Assuming Meaning&lt;/em&gt;&lt;/p&gt;&lt;/blockquote&gt;&#xA;&lt;h2 id=&#34;executive-summary-from-symbol-to-system&#34;&gt;&lt;strong&gt;Executive Summary: From Symbol to System&lt;/strong&gt;&lt;/h2&gt;&#xA;&lt;blockquote&gt;&#xA;&lt;p&gt;&lt;em&gt;We set out to understand a single image. We ended up building a system that can understand structure itself.&lt;/em&gt;&lt;/p&gt;&lt;/blockquote&gt;&#xA;&lt;p&gt;We started with a constraint:&lt;/p&gt;&#xA;&lt;blockquote&gt;&#xA;&lt;p&gt;&lt;strong&gt;Assume we do not understand the symbol.&lt;/strong&gt;&lt;/p&gt;&lt;/blockquote&gt;&#xA;&lt;p&gt;No prior knowledge. No accepted interpretations.&lt;/p&gt;&#xA;&lt;p&gt;Just an image: an eye, a triangle, rays, an unfinished pyramid.&lt;/p&gt;&#xA;&lt;p&gt;From that starting point, we:&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
