Skip to content

Research

Principled machine learning, grounded in the life sciences

We develop machine learning methods that are reliable, data-efficient and aware of their own uncertainty, and we apply them to challenging problems in biology, agriculture, ecology and medicine.

01

Uncertainty quantification & probabilistic models

This research line started more than ten years ago, at that stage with a focus on Bayes-optimal inference from probabilistic models. Since then our interests have extended to quantifying epistemic and aleatoric uncertainty, classifier calibration, set-valued prediction, prediction intervals, conformal prediction, open-set recognition and out-of-distribution detection.

  • Epistemic & aleatoric uncertainty
  • Calibration
  • Conformal prediction
  • Set-valued prediction
  • OOD detection
Key publication
02

Multi-target prediction

Multi-target prediction is an umbrella term for complex machine learning settings such as multi-label classification, multivariate regression, pairwise learning, dyadic prediction, matrix completion and zero-shot learning. Willem Waegeman has co-organized many tutorials and workshops on the topic and published numerous papers on it. A current focus is the development of flexible AutoML software for multi-target prediction problems.

  • Multi-label classification
  • Pairwise learning
  • Matrix completion
  • Zero-shot learning
  • AutoML
Key publication
03

Applications in the life sciences

Over the last decade we have developed machine learning solutions for all sorts of -omics data — genomics, epigenomics, transcriptomics and proteomics — as well as for specific phenotyping technologies, with applications in environmental sciences, agriculture and medicine. Larger projects also involved the analysis of climate time series and timber provenance.

  • Genomics
  • Transcriptomics
  • Proteomics
  • Agriculture
  • Ecology
Key publication
04

Sequence models & time series analysis

Specific interests include sequence labelling, autoregressive models, transformers for biological sequences, time series forecasting, causal inference in time series, feature learning for sequential data and spatio-temporal problems.

  • Transformers
  • Sequence labelling
  • Forecasting
  • Granger causality
  • Spatio-temporal models
Key publication

Impact

Most cited work

Citation counts from Google Scholar, as of September 2026.

All publications
  1. 3517 cites

    Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods

    Eyke Hüllermeier, Willem Waegeman

    Machine Learning · 2021

  2. 930 cites

    A community effort to assess and improve drug sensitivity prediction algorithms

    James C Costello, Laura M Heiser, Elisabeth Georgii, Mehmet Gönen et al.

    Nature biotechnology · 2014

  3. 557 cites

    On label dependence and loss minimization in multi-label classification

    Krzysztof Dembczyński, Willem Waegeman, Weiwei Cheng, Eyke Hüllermeier

    Machine Learning · 2012

  4. 459 cites

    Absolute quantification of microbial taxon abundances

    Ruben Props, Frederiek-Maarten Kerckhof, Peter Rubbens, Jo De Vrieze et al.

    The ISME journal · 2017

  5. 300 cites

    Bacterial species identification from MALDI-TOF mass spectra through data analysis and machine learning

    Katrien De Bruyne, Bram Slabbinck, Willem Waegeman, Paul Vauterin et al.

    Systematic and applied microbiology · 2011

  6. 240 cites

    An experimental comparison of cross-validation techniques for estimating the area under the ROC curve

    Antti Airola, Tapio Pahikkala, Willem Waegeman, Bernard De Baets et al.

    Computational Statistics & Data Analysis · 2011