AIConfPaper
← Search

Willem Waegeman

7 accepted papers

2026

Position: Agentic AI systems should be making Bayes-consistent decisions

Theodore Papamarkou, Pierre Alquier, Matthias Bauer, Wray Buntine, Andrew Davison, Gintare Karolina Dziugaite +24
ICML 2026poster

LLMs excel at predictive tasks and complex reasoning tasks, but many high-value deployments rely on decisions under uncertainty, for example, which tool to call, which expert to consult, or how many resources to invest. While the usefulness and feasibility of Bayesian approaches remain unclear for L…

Cited by 0Source ↗Scholar ↗
2026

Position: Epistemic uncertainty estimation methods are fundamentally incomplete

Sebastian Jimenez, Mira Juergens, Willem Waegeman
ICML 2026poster

Identifying and disentangling sources of predictive uncertainty is essential for trustworthy supervised learning. We argue that widely used second-order decomposition-based approaches to uncertainty quantification are fundamentally incomplete. First, we show that unaccounted bias contaminates uncert…

Cited by 0Source ↗Scholar ↗
2024

Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?

Mira Juergens, Nis Meinert, Viktor Bengs, Eyke Hüllermeier, Willem Waegeman
ICML 2024poster

Trustworthy ML systems should not only return accurate predictions, but also a reliable representation of their uncertainty. Bayesian methods are commonly used to quantify both aleatoric and epistemic uncertainty, but alternative approaches, such as evidential deep learning methods, have become popu…

Cited by 10Source ↗PDF ↗code ↗Scholar ↗
2023

On Second-Order Scoring Rules for Epistemic Uncertainty Quantification

Viktor Bengs, Eyke Hüllermeier, Willem Waegeman
ICML 2023poster

It is well known that accurate probabilistic predictors can be trained through empirical risk minimisation with proper scoring rules as loss functions. While such learners capture so-called aleatoric uncertainty of predictions, various machine learning methods have recently been developed with the g…

Cited by 39Source ↗PDF ↗Scholar ↗
2023

On the Calibration of Probabilistic Classifier Sets

Thomas Mortier, Viktor Bengs, Eyke Hüllermeier, Stijn Luca, Willem Waegeman
AISTATS 2023poster

Multi-class classification methods that produce sets of probabilistic classifiers, such as ensemble learning methods, are able to model aleatoric and epistemic uncertainty. Aleatoric uncertainty is then typically quantified via the Bayes error, and epistemic uncertainty via the size of the set. In t…

Cited by 11Source ↗PDF ↗Scholar ↗
2022

Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisation

Viktor Bengs, Eyke Hüllermeier, Willem Waegeman
NeurIPS 2022accept

Uncertainty quantification has received increasing attention in machine learning in the recent past. In particular, a distinction between aleatoric and epistemic uncertainty has been found useful in this regard. The latter refers to the learner's (lack of) knowledge and appears to be especially diff…

Cited by 50Source ↗PDF ↗Scholar ↗
2022

Set-valued prediction in hierarchical classification with constrained representation complexity

Thomas Mortier, Eyke Hüllermeier, Krzysztof Dembczyński, Willem Waegeman
UAI 2022poster

Set-valued prediction is a well-known concept in multi-class classification. When a classifier is uncertain about the class label for a test instance, it can predict a set of classes instead of a single class. In this paper, we focus on hierarchical multi-class classification problems, where valid s…

Cited by 4Source ↗PDF ↗Scholar ↗