<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Martingale on Yuchen Du</title><link>https://dudusoar.github.io/tags/martingale/</link><description>Recent content in Martingale on Yuchen Du</description><generator>Hugo -- 0.164.0</generator><language>en-us</language><lastBuildDate>Fri, 17 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://dudusoar.github.io/tags/martingale/index.xml" rel="self" type="application/rss+xml"/><item><title>Square the Error to Turn Convergence into Energy Dissipation</title><link>https://dudusoar.github.io/posts/stochastic-approximation/16-square-error-as-energy/</link><pubDate>Fri, 17 Jul 2026 00:00:00 +0000</pubDate><guid>https://dudusoar.github.io/posts/stochastic-approximation/16-square-error-as-energy/</guid><description>The Dvoretzky proof tracks squared error, shows that its positive drift is summable, and rules out a positive limiting energy.</description></item><item><title>Two Proof Routes Explain Different Levels of Convergence</title><link>https://dudusoar.github.io/posts/stochastic-approximation/22-two-proof-routes-to-dvoretzky-convergence/</link><pubDate>Fri, 17 Jul 2026 00:00:00 +0000</pubDate><guid>https://dudusoar.github.io/posts/stochastic-approximation/22-two-proof-routes-to-dvoretzky-convergence/</guid><description>Expectation and variance explain distributional concentration; the energy and martingale route proves almost-sure convergence.</description></item></channel></rss>