Hello there!

Hello, I am a happy postdoc at Linköping University (Linköping, Sweden). My main area of interest encompasses unsupervised learning, specifically discriminative clustering, and its application with or without constraints for instance in health or material science.

I am currently focusing on the prediction of synthesisability or dynamic stability of MXenes materials from MAX phases using ML. This type of problen is tightly linked to unsupervised methods, as it covers both PU learning (binary classification where class observation is missing) and/or semi-supervised clustering (like semi-supervised learning, but with veerrrryyy little annotated data). I work closely with Fredrik Lindsten. Do not hesitate to contact me for whatever research-related topic.

I am also responsible of the ML seminar series at the department of computer science (IDA) of the Linköping University.

Background

I graduated from a Ph.D. in computer science at the Université Côte d’Azur (Nice, France) and a cotutelle Ph.D. in molecular medicine at the Université Laval (Quebec, Canada). The topic was the qualification of the aortic stenosis, a severe cardiac disease, throughout discriminative clustering methods. I had the pleasure of being supervised by Frédéric Precioso, Pierre-Alexandre Mattei in France and Arnaud Droit in Canada. If you need a quick insight on my work, you can check the GemClus package that summarises most of my PhD algorithms. Prior to academia, I graduated within computer science with a double degree: an engineering one at the Institut National des Sciences Appliquées de Lyon and a Master’s one at KTH Royal Institute of Technology.

Some keywords:

Discriminative clustering / Unsupervised learning / Decision trees / Ensemble methods / AI for health / AI for material science

news

Jun 10, 2026	Have been selected as a top reviewer for UAI 2026, that’s cool and many thanks!
Dec 02, 2025	I’ll be at the ELLIS UnConference 2025 at Copenhagen presenting at the Poster Session my paper “Discrimintative Ordering through Ensemble Consensus” (DISCOTEC).
Nov 26, 2025	Hugo Sénétaire and I will present a poster titled “Your Contrastive Learning Model is Secretly an EBM and You Should Treat it Like One” at the Nordic AI Meet 2025 in Norrköping.
Oct 13, 2025	I have been selected as a NeurIPS top reviewer. How cool!
Jul 18, 2025	Will present a paper at the UAI 2025 conference at Rio de Janeiro on a novel score for clustering evaluation, called DISCOTEC.

selected publications

ACM Comp. Sur.
A Tutorial on Discriminative Clustering and Mutual Information

Louis Ohl, Pierre-Alexandre Mattei, and Frederic Precioso

ACM Comput. Surv., Oct 2025

Place: New York, NY, USA

Abs DOI Bib HTML Code

To cluster data is to separate samples into distinctive groups that should ideally have some cohesive properties. Today, numerous clustering algorithms exist, and their differences lie essentially in what can be perceived as “cohesive properties”. Therefore, hypotheses on the nature of clusters must be set: they can be either generative or discriminative. As the last decade witnessed the impressive growth of deep clustering methods that involve neural networks to handle high-dimensional data often in a discriminative manner; we concentrate mainly on the discriminative hypotheses. In this article, our aim is to provide an accessible historical perspective on the evolution of discriminative clustering methods and notably how the nature of assumptions of the discriminative models changed over time: from decision boundaries to invariance critics. We notably highlight how mutual information has been a historical cornerstone of the progress of (deep) discriminative clustering methods. We also show some known limitations of mutual information and how discriminative clustering methods tried to circumvent those. We then discuss the challenges that discriminative clustering faces with respect to the selection of the number of clusters. Finally, we showcase these techniques using the dedicated Python package, GemClus , that we have developed for discriminative clustering.
@article{ohl_tutorial_2025, title = {A {Tutorial} on {Discriminative} {Clustering} and {Mutual} {Information}}, volume = {58}, issn = {0360-0300}, url = {https://doi.org/10.1145/3748255}, doi = {10.1145/3748255}, number = {4}, journal = {ACM Comput. Surv.}, publisher = {Association for Computing Machinery}, author = {Ohl, Louis and Mattei, Pierre-Alexandre and Precioso, Frederic}, month = oct, year = {2025}, note = {Place: New York, NY, USA}, keywords = {Clustering, discriminative models, mutual information, neural networks, unsupervised learning}, pages = {1--36}, }