Scholar
Bálint Máté
Google Scholar ID: Uhcp1DAAAAAJ
PhD student, University of Geneva
Citations & Impact
Citations
75
H-index
4
i10-index
2
Publications
11
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- [{'title': 'Solvation Free Energies from Neural Thermodynamic Integration', 'year': 2025, 'authors': 'B. Máté, F. Fleuret and T. Bereau', 'journal': "The Journal of Chemical Physics 162 (12), Editor's Pick"}, {'title': 'Neural Thermodynamic Integration: Free Energies from Energy-based Diffusion Models', 'year': 2024, 'authors': 'B. Máté, F. Fleuret and T. Bereau', 'journal': 'The Journal of Physical Chemistry Letters 15 (45)'}, {'title': 'Multi-Lattice Sampling of Quantum Field Theories via Neural Operator-based Flows', 'year': 2024, 'authors': 'B. Máté and F. Fleuret', 'journal': 'Machine Learning: Science and Technology 5 (4)'}, {'title': 'Learning Interpolations between Boltzmann Densities', 'year': 2023, 'authors': 'B. Máté and F. Fleuret', 'journal': 'Transactions on Machine Learning Research (TMLR)'}, {'title': 'Flowification: Everything is a Normalizing Flow', 'year': 2022, 'authors': 'B. Máté, S. Klein, T. Golling and F. Fleuret', 'journal': 'Neural Information Processing Systems (NeurIPS)'}]
Research Experience
- Interned at the Chemistry team of Meta FAIR working on organic crystal structure prediction; visited Tristan Bereau's group to learn about free energies; and spent some time with the AI4Science group at Microsoft Research, contributing to their effort to machine learn the exchange-correlation functional.
Education
- Pursuing his PhD under the supervision of François Fleuret; previously studied theoretical physics and differential geometry in Hamburg and mechanical engineering in Budapest.
Background
- Doctoral student in Computer Science and Physics, focusing on generative modeling, sampling methods, and free-energy estimation.
