
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>vishalv.com</title>
      <link>https://www.vishalv.com/tags/slow-weights</link>
      <description>Personal website of Vishal V, featuring writing, ideas, projects, and research</description>
      <language>en-us</language>
      <managingEditor>vishalvignesh.iitm@gmail.com (Vishal V)</managingEditor>
      <webMaster>vishalvignesh.iitm@gmail.com (Vishal V)</webMaster>
      <lastBuildDate>Mon, 31 Aug 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://www.vishalv.com/tags/slow-weights/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://www.vishalv.com/notes/schlagLinearTransformersAre2021</guid>
    <title>Linear Transformers Are Secretly Fast Weight Programmers</title>
    <link>https://www.vishalv.com/notes/schlagLinearTransformersAre2021</link>
    <description>We show the formal equivalence of linearised self-attention mechanisms and fast weight controllers from the early ’90s, where a slow neural net learns by gradient descent to program the fast weights of another net through sequences of elementary programming instructions which are...</description>
    <pubDate>Mon, 31 Aug 2026 00:00:00 GMT</pubDate>
    <author>vishalvignesh.iitm@gmail.com (Vishal V)</author>
    <category>self-attention</category><category>linear-transformer</category><category>delta-rule</category><category>fast-weights</category><category>slow-weights</category><category>outer-product</category>
  </item>

    </channel>
  </rss>
