<?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>Conformal-Prediction on The Standard Error</title><link>https://jongha-jeon-dev.github.io/standarderror/tags/conformal-prediction/</link><description>Recent content in Conformal-Prediction on The Standard Error</description><generator>Hugo -- 0.123.7</generator><language>en-gb</language><lastBuildDate>Thu, 08 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://jongha-jeon-dev.github.io/standarderror/tags/conformal-prediction/index.xml" rel="self" type="application/rss+xml"/><item><title>Coverage 1: 90% Coverage Is a Promise About Averages, Not About You</title><link>https://jongha-jeon-dev.github.io/standarderror/lectures/coverage-1-marginal/</link><pubDate>Fri, 11 Sep 2026 00:00:00 +0000</pubDate><guid>https://jongha-jeon-dev.github.io/standarderror/lectures/coverage-1-marginal/</guid><description>Conformal prediction&amp;#39;s guarantee is exact, distribution-free and finite-sample, and it is an average over the test distribution. Conditioned on the model&amp;#39;s own confidence it runs from 82% to 95% - and then it turns out that the unevenness is the score&amp;#39;s fault, not the method&amp;#39;s.</description></item><item><title>Coverage 5: The Split You Chose Is Hiding How Variable Your Coverage Is</title><link>https://jongha-jeon-dev.github.io/standarderror/lectures/coverage-5-split/</link><pubDate>Thu, 08 Oct 2026 00:00:00 +0000</pubDate><guid>https://jongha-jeon-dev.github.io/standarderror/lectures/coverage-5-split/</guid><description>Episode 1 promised an error bar wider than exchangeability allows. Most of that was my formula. What is left appears only when you split the way a deployment does.</description></item></channel></rss>