Hi! I'm Bálint, I work as an ML researcher at Peptone.

I obtained my doctorate in computer science and physics focusing on generative modeling, sampling and free-energy estimation. My supervisor was François Fleuret. During my PhD I had the chance to intern with the chemistry team of FAIR at Meta working on organic crystal structure prediction; to visit the group of Tristan Bereau where I learnt about free energies; and to spend some time in the AI4Science group of Microsoft Research where I played a small part in their effort to machine learn the exchange-correlation functional.

Prior to starting my PhD, I studied theoretical physics and differential geometry in Hamburg and mechanical engineering in Budapest.

[scholar] [github] [twitter] [email]



Publications
Solvation Free Energies from Neural Thermodynamic Integration
2025
B. Máté, F. Fleuret and T. Bereau
The Journal of Chemical Physics 162 (12), Editor's Pick
[pdf] [summary] [bib]
Neural Thermodynamic Integration: Free Energies from Energy-based Diffusion Models
2024
B. Máté, F. Fleuret and T. Bereau
The Journal of Physical Chemistry Letters 15 (45)
[pdf] [summary] [bib]
Multi-Lattice Sampling of Quantum Field Theories via Neural Operator-based Flows
2024
B. Máté and F. Fleuret
Machine Learning: Science and Technology 5 (4)
[pdf] [bib]
Learning Interpolations between Boltzmann Densities
2023
B. Máté and F. Fleuret
Transactions on Machine Learning Research (TMLR)
[pdf] [summary] [bib]
Flowification: Everything is a Normalizing Flow
2022
B. Máté, S. Klein, T. Golling and F. Fleuret
Neural Information Processing Systems (NeurIPS)
[pdf] [bib]