Atamert Rahma, M.Sc.
Technical University of Munich
TUM School of CIT
Department of Computer Science
Boltzmannstrasse 3
85748 Garching
Germany
Office: MI 02.05.041
Mail: atamert.rahma (at) tum.de
Office Hours: by arrangement
Links: Google Scholar ResearchGate GitHub Blog
Background
- Started my doctoral studies with Prof. Dr. Felix Dietrich in December 2024
- M.Sc. in Informatics 2021-2024
- B.Sc. in Informatics 2018-2021
- From Istanbul, Türkiye
Research interests
- Accelarating training of neural networks using randomized algorithms
- Approximating physical systems (e.g. Hamiltonian) using machine learning
- Accelerating numerical simulations (e.g. molecular dynamics) using machine learning methods
Thesis supervision
Please send me an email if you are interested in doing a thesis, guided research, or IDP with me. I can supervise students from across CIT. My current list of topics (might be outdated / regularly updated) is
- Accelerating reinforcement learning using random features
- Coarse-graining molecular dynamics simulations using machine learning
- Solving sparse and ill-conditioned linear systems
Bachelor thesis
Baris Ozgun: Benchmark Analysis of Random Feature Networks for Learning Interatomic Potentials. 2026 (mediaTUM)
Sofia Koufogazou Loukianova: Random Feature Hamiltonian Networks for N-Body Problems. 2026 (mediaTUM)
Cagan Akin: Learning Hamiltonian Functions of Handwritten Digits using Neural Networks. 2025 (mediaTUM)
Frederik Maximilian Willemsen: Data-Driven Selection of Algorithmic Configurations in Particle Simulations. 2025 (mediaTUM)
Master thesis
Deniz Erdogan: Structure-Preserving Modeling of Non-Conservative Systems via Random Feature Hamiltonian Networks. 2026 (mediaTUM)
Ioan-Daniel Craciun: Conditional diffusion models for molecular dynamics conformation sampling. 2025 (mediaTUM)