Research

My PhD research is about uncertainty quantification in machine learning, with a focus on the question of how to evaluate and disentangle representations of aleatoric and epistemic uncertainty. I am also interested in uncertainty-aware prediction pipelines like conformal and selective prediction.

When can you trust the annotation? Selective Prediction for Molecular Structure Retrieval from MS/MS Spectra

We propose a selective prediction framework for the task of molecular structure retrieval via molecular fingerprint prediction from mass spectra..

Bias contamination in uncertainty estimates

Epistemic uncertainty estimation methods are fundamentally incomplete

We argue that current epistemic uncertainty methods are fundamentally incomplete: unaccounted bias contaminates aleatoric estimates, and existing methods capture only partial variance contributions.

A calibration test for evaluating set-based epistemic uncertainty representations

We propose a nonparametric hypothesis test for evaluating set-based epistemic uncertainty representations.

Faithfulness violations in evidential deep learning

Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?

We show that evidential deep learning methods face fundamental optimisation difficulties and that their epistemic uncertainty estimates do not satisfy basic faithfulness properties.