<?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>Uncertainty Quantification on Mira Jürgens</title><link>https://mkjuergens.github.io/tags/uncertainty-quantification/</link><description>Recent content in Uncertainty Quantification on Mira Jürgens</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Wed, 01 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://mkjuergens.github.io/tags/uncertainty-quantification/index.xml" rel="self" type="application/rss+xml"/><item><title>When can you trust the annotation? Selective Prediction for Molecular Structure Retrieval from MS/MS Spectra</title><link>https://mkjuergens.github.io/research/msms/</link><pubDate>Wed, 01 Apr 2026 00:00:00 +0000</pubDate><guid>https://mkjuergens.github.io/research/msms/</guid><description>We propose a selective prediction framework for the task of molecular structure retrieval via molecular fingerprint prediction from mass spectra.</description></item><item><title>Epistemic uncertainty estimation methods are fundamentally incomplete</title><link>https://mkjuergens.github.io/research/position/</link><pubDate>Fri, 06 Feb 2026 00:00:00 +0000</pubDate><guid>https://mkjuergens.github.io/research/position/</guid><description>We argue that current epistemic uncertainty methods are fundamentally incomplete: unaccounted bias contaminates aleatoric estimates, and existing methods capture only partial variance contributions. — Machine Learning (Springer), 2025.</description></item><item><title>Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?</title><link>https://mkjuergens.github.io/research/edl/</link><pubDate>Sun, 21 Jul 2024 00:00:00 +0000</pubDate><guid>https://mkjuergens.github.io/research/edl/</guid><description>We show that evidential deep learning methods face fundamental optimisation difficulties and that their epistemic uncertainty estimates do not satisfy basic faithfulness properties. — ICML, 2024.</description></item></channel></rss>