<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Home</title><description>Conor Heins — researcher and engineer working on Bayesian machine learning, active inference, and model-based reinforcement learning.</description><link>https://conorheins.github.io/</link><item><title>AXIOM</title><link>https://conorheins.github.io/projects/axiom/</link><guid isPermaLink="true">https://conorheins.github.io/projects/axiom/</guid><description>Object-centric world models that learn to play games in minutes by expanding their structure online.</description><pubDate>Sat, 31 May 2025 00:00:00 GMT</pubDate></item><item><title>CAVI-CMN: Gradient-Free Variational Learning</title><link>https://conorheins.github.io/projects/cavi-cmn/</link><guid isPermaLink="true">https://conorheins.github.io/projects/cavi-cmn/</guid><description>Closed-form coordinate-ascent variational inference for conditional mixture networks — fast, exact updates, no gradients.</description><pubDate>Thu, 29 Aug 2024 00:00:00 GMT</pubDate></item><item><title>Collective Behavior from Surprise Minimization</title><link>https://conorheins.github.io/projects/collective-behavior/</link><guid isPermaLink="true">https://conorheins.github.io/projects/collective-behavior/</guid><description>The first active inference model of collective motion — surprise-minimizing agents reproduce real-world flocking patterns.</description><pubDate>Mon, 22 Apr 2024 00:00:00 GMT</pubDate></item><item><title>pymdp</title><link>https://conorheins.github.io/projects/pymdp/</link><guid isPermaLink="true">https://conorheins.github.io/projects/pymdp/</guid><description>A Python library for active inference in discrete state-spaces. 680+ GitHub stars.</description><pubDate>Fri, 14 Jan 2022 00:00:00 GMT</pubDate></item></channel></rss>