26.07.2023

Teaser image to

MCML Researchers With Three Papers at UAI 2023

39th Conference on Uncertainty in Artificial Intelligence (UAI 2023). Pittsburgh, PA, USA, 31.07.2023–04.08.2023

We are happy to announce that MCML researchers are represented with three papers at UAI 2023. Congrats to our researchers!

Main Track (3 papers)

J. Rodemann, J. Goschenhofer, E. Dorigatti, T. Nagler and T. Augustin.
Approximately Bayes-optimal pseudo-label selection.
UAI 2023 - 39th Conference on Uncertainty in Artificial Intelligence. Pittsburgh, PA, USA, Jul 31-Aug 03, 2023. URL
Abstract

Semi-supervised learning by self-training heavily relies on pseudo-label selection (PLS). This selection often depends on the initial model fit on labeled data. Early overfitting might thus be propagated to the final model by selecting instances with overconfident but erroneous predictions, often referred to as confirmation bias. This paper introduces BPLS, a Bayesian framework for PLS that aims to mitigate this issue. At its core lies a criterion for selecting instances to label: an analytical approximation of the posterior predictive of pseudo-samples. We derive this selection criterion by proving Bayes-optimality of the posterior predictive of pseudo-samples. We further overcome computational hurdles by approximating the criterion analytically. Its relation to the marginal likelihood allows us to come up with an approximation based on Laplace’s method and the Gaussian integral. We empirically assess BPLS on simulated and real-world data. When faced with high-dimensional data prone to overfitting, BPLS outperforms traditional PLS methods.

MCML Authors
Link to Profile Thomas Nagler

Thomas Nagler

Prof. Dr.

Computational Statistics & Data Science


Y. Sale, M. Caprio and E. Hüllermeier.
Is the Volume of a Credal Set a Good Measure for Epistemic Uncertainty?
UAI 2023 - 39th Conference on Uncertainty in Artificial Intelligence. Pittsburgh, PA, USA, Jul 31-Aug 03, 2023. URL
Abstract

Adequate uncertainty representation and quantification have become imperative in various scientific disciplines, especially in machine learning and artificial intelligence. As an alternative to representing uncertainty via one single probability measure, we consider credal sets (convex sets of probability measures). The geometric representation of credal sets as d-dimensional polytopes implies a geometric intuition about (epistemic) uncertainty. In this paper, we show that the volume of the geometric representation of a credal set is a meaningful measure of epistemic uncertainty in the case of binary classification, but less so for multi-class classification. Our theoretical findings highlight the crucial role of specifying and employing uncertainty measures in machine learning in an appropriate way, and for being aware of possible pitfalls.

MCML Authors
Link to Profile Eyke Hüllermeier

Eyke Hüllermeier

Prof. Dr.

Artificial Intelligence and Machine Learning


L. Wimmer, Y. Sale, P. Hofman, B. Bischl and E. Hüllermeier.
Quantifying Aleatoric and Epistemic Uncertainty in Machine Learning: Are Conditional Entropy and Mutual Information Appropriate Measures?
UAI 2023 - 39th Conference on Uncertainty in Artificial Intelligence. Pittsburgh, PA, USA, Jul 31-Aug 03, 2023. URL
Abstract

The quantification of aleatoric and epistemic uncertainty in terms of conditional entropy and mutual information, respectively, has recently become quite common in machine learning. While the properties of these measures, which are rooted in information theory, seem appealing at first glance, we identify various incoherencies that call their appropriateness into question. In addition to the measures themselves, we critically discuss the idea of an additive decomposition of total uncertainty into its aleatoric and epistemic constituents. Experiments across different computer vision tasks support our theoretical findings and raise concerns about current practice in uncertainty quantification.

MCML Authors
Link to website

Lisa Wimmer

Statistical Learning and Data Science

Link to website

Paul Hofman

Artificial Intelligence and Machine Learning

Link to Profile Bernd Bischl

Bernd Bischl

Prof. Dr.

Statistical Learning and Data Science

Link to Profile Eyke Hüllermeier

Eyke Hüllermeier

Prof. Dr.

Artificial Intelligence and Machine Learning


26.07.2023


Subscribe to RSS News feed

Related

Link to When Clinical Expertise Meets AI Innovation – with Michael Ingrisch

25.06.2025

When Clinical Expertise Meets AI Innovation – With Michael Ingrisch

The new research film features Michael Ingrisch, who shows how AI and clinical expertise can solve real challenges in radiology together.

Link to Autonomous Driving: From Infinite Possibilities to Safe Decisions— with Matthias Althoff

23.06.2025

Autonomous Driving: From Infinite Possibilities to Safe Decisions— With Matthias Althoff

The new research film features Matthias Althoff explaining how his team verifies autonomous vehicle safety using EDGAR and rigorous testing.

Link to ERC Advanced Grant for Massimo Fornasier

20.06.2025

ERC Advanced Grant for Massimo Fornasier

Massimo Fornasier was awarded ERC Advanced Grant to develop advanced algorithms for solving complex nonconvex optimization problems.

Link to ERC Advanced Grant for Albrecht Schmidt

18.06.2025

ERC Advanced Grant for Albrecht Schmidt

Albrecht Schmidt receives ERC Advanced Grant for research on personalized generative AI to support memory, planning, and creativity.

Link to Better Data, Smarter AI: Why Quality Matters – with Frauke Kreuter

11.06.2025

Better Data, Smarter AI: Why Quality Matters – With Frauke Kreuter

In our new research film, Frauke Kreuter explains how data quality shapes fair, reliable, and socially responsible AI systems.