Bálint Mucsányi
PhD Candidate in Machine Learning
University of Tübingen
About Me
I am an ELLIS and IMPRS-IS PhD candidate in Machine Learning at the University of Tübingen, advised by Philipp Hennig, Bernhard Schölkopf, and Yee Whye Teh. My research focuses on uncertainty quantification and Bayesian deep learning. I am particularly excited about the parameter spaces of neural networks and symmetries therein.
Previously, I completed my Master's degree with distinction at the University of Tübingen in the STAI lab of Seong Joon Oh and worked as a research intern at the Mackelab. I received my Bachelor's degree with the Best Thesis and Outstanding Student of the Faculty awards from ELTE Eötvös Loránd University.
Education
PhD in Machine Learning
University of Tübingen, 2024 - Present
MSc in Machine Learning
University of Tübingen, 2021 - 2024
BSc in Computer Science
ELTE Eötvös Loránd University, 2018 - 2021
Recent News
January 2025
Our paper "Debiasing Mini-Batch Quadratics for Applications in Deep Learning" has been accepted to ICLR 2025!
September 2024
Our paper "Benchmarking Uncertainty Disentanglement: Specialized Uncertainties for Specialized Tasks" has been accepted to NeurIPS 2024 as a Spotlight!
July 2024
I started my PhD in Machine Learning! I will be working on uncertainty quantification and Bayesian deep learning. I am advised by Philipp Hennig, Bernhard Schölkopf, and Yee Whye Teh.
March 2024
I graduated with distinction from the Machine Learning Master's of the University of Tübingen.
Publications
Rethinking Approximate Gaussian Inference in Classification
Bálint Mucsányi, Nathaël Da Costa, Philipp Hennig
arXiv, 2025
Debiasing Mini-Batch Quadratics for Applications in Deep Learning
Lukas Tatzel, Bálint Mucsányi, Osane Hackel, Philipp Hennig
ICLR, 2025
sbi Reloaded: A Toolkit for Simulation-Based Inference Workflows
Jan Boelts, Michael Deistler, Manuel Gloeckler, Álvaro Tejero-Cantero, Jan-Matthis Lueckmann, Guy Moss, Peter Steinbach, Thomas Moreau, Fabio Muratore, Julia Linhart, Conor Durkan, Julius Vetter, Benjamin Kurt Miller and Maternus Herold, Abolfazl Ziaeemehr, Matthijs Pals, Theo Gruner, Sebastian Bischoff, Nastya Krouglova, Richard Gao, Janne K. Lappalainen, Bálint Mucsányi, Felix Pei and Auguste Schulz, Zinovia Stefanidi, Pedro Rodrigues and Cornelius Schröder, Faried Abu Zaid, Jonas Beck and Jaivardhan Kapoor, David S. Greenberg, Pedro J. Gonçalves and Jakob H. Macke
arXiv, 2024
Trustworthy Machine Learning
Bálint Mucsányi, Michael Kirchhof, Elisa Nguyen and Alexander Rubinstein, Seong Joon Oh
arXiv, 2023