<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0">
    <channel>
      <title>Category Theory Wiki</title>
      <link>https://mathstruct.org/CategoryTheory-ML-Wiki</link>
      <description>Last 10 notes on Category Theory Wiki</description>
      <generator>Quartz -- quartz.jzhao.xyz</generator>
      <item>
    <title>New Insights and Perspectives on the Natural Gradient Method</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Papers/New-Insights-and-Perspectives-on-the-Natural-Gradient-Method</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Papers/New-Insights-and-Perspectives-on-the-Natural-Gradient-Method</guid>
    <description><![CDATA[ paper New Insights and Perspectives on the Natural Gradient Method — James Martens (2014). ]]></description>
    <pubDate>Sat, 24 Oct 2026 14:43:21 GMT</pubDate>
  </item><item>
    <title>Automatic Differentiation of Expected Values</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Automatic-Differentiation-of-Expected-Values</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Automatic-Differentiation-of-Expected-Values</guid>
    <description><![CDATA[ definition theorem example program Many objectives in machine learning are expected values of random processes: an expected reward over simulated runs, a loss over random minibatches, an ELBO over samples of a variational family. ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Diffeological Space</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Diffeological-Space</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Diffeological-Space</guid>
    <description><![CDATA[ definition theorem example A diffeological space (Baez &amp; Hoffnung, Definition 2, after Souriau) is a set X with, for every open U \subseteq \mathbb R^n, a set of functions \varphi : U \to X called plots, such that a plot precomposed with a smooth map f : U&#039; \to U of open sets is a plot (smo... ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Entropy in a Markov Category</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Entropy-in-a-Markov-Category</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Entropy-in-a-Markov-Category</guid>
    <description><![CDATA[ definition theorem example Perrone measures information in a Markov Category by how far an equation fails to hold. ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Fisher Information Metric</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Fisher-Information-Metric</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Fisher-Information-Metric</guid>
    <description><![CDATA[ definition theorem example For a smooth family of probability distributions p_\theta on \Omega, parametrised by \theta \in \Theta \subseteq \mathbb R^D, the score is the gradient of the log-likelihood, s_\theta(\omega) = \nabla_\theta \log p_\theta(\omega), and the Fisher information matrix is its s... ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Giry Monad</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Giry-Monad</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Giry-Monad</guid>
    <description><![CDATA[ definition theorem example The Giry monad \mathcal G on the category \mathbf{Meas} of measurable spaces sends X to the space \mathcal G X of probability measures on X, with the \sigma-algebra generated by the evaluation maps \mu \mapsto \mu(A). ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Information Geometry of Probabilistic Programs</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Information-Geometry-of-Probabilistic-Programs</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Information-Geometry-of-Probabilistic-Programs</guid>
    <description><![CDATA[ definition theorem example Karwowski, Nielsen &amp; Staton give probabilistic programs a compositional information geometry. ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Map of Content</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Map-of-Content</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Map-of-Content</guid>
    <description><![CDATA[ index annotation New here? Read Start Here first. ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Markov Category</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Markov-Category</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Markov-Category</guid>
    <description><![CDATA[ definition theorem example A Markov category (Fritz, Definition 2.1) is a Symmetric Monoidal Category in which every object X carries a commutative comonoid \mathrm{copy}_X : X \to X \otimes X, \mathrm{del}_X : X \to I, compatible with \otimes, such that deletion is natural: f \mathbin{;} \mathrm{de... ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item><item>
    <title>Natural Gradient</title>
    <link>https://mathstruct.org/CategoryTheory-ML-Wiki/Natural-Gradient</link>
    <guid>https://mathstruct.org/CategoryTheory-ML-Wiki/Natural-Gradient</guid>
    <description><![CDATA[ definition theorem example The natural gradient of a loss \ell(\theta) on a statistical model is the ordinary gradient with its index raised by the Fisher Information Metric I(\theta) (Amari 1998; Nielsen eq. ]]></description>
    <pubDate>Thu, 22 Oct 2026 18:39:47 GMT</pubDate>
  </item>
    </channel>
  </rss>