Large scale computing has transformed scientific discovery both through physical simulation and through machine learning. Yet these two forms of scientific computing sit in uneasy relationship, operating with distinct logic and methodology.
This semester we will investigate the interface between machine learning and simulation. We’ll look at the underlying structure of simulator models, and learn about energy-based, implicit and non-differentiable models. We’ll discuss emerging ideas about how to do inference on simulators and how to interface with experimental data. We’ll focus on the context of molecular dynamics and other simulation methods common in the molecular sciences. Overall, the aim is to provide a background for researchers interested in advancing the frontier of ML and simulation.
Thursdays 1-2:30pm, Building 207 Room 222 (DTU Chemistry). First meeting: February 5.
Please sign up here for the mailing list (scheduling announcements, etc.)
Reading list:
- Feb. 5: Sampling from energy-based models.
Noé et al. 2019. Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning. https://www.science.org/doi/10.1126/science.aaw1147
Tentative future papers/topics include
Learning in implicit generative models, Mohamed and Lakshminarayanan. https://arxiv.org/abs/1610.03483
Do Differentiable Simulators Give Better Policy Gradients? Suh et al. 2022. https://proceedings.mlr.press/v162/suh22b.html
Ab initio solution of the many-electron Schrödinger equation with deep neural networks. Pfau et al. 2020
Fourier Neural Operators Li et al. 2021. https://arxiv.org/pdf/2010.08895
About ML & Molecules reading group: ML & Molecules is a reading group organized by Asst. Prof. Eli Weinstein (DTU Chemistry) and Assoc. Prof. Jes Frellsen (DTU Compute). It focuses on fundamental probabilistic machine learning and its intersection with the molecular sciences. Everyone in the group presents a paper, in a rotating schedule. It is open to students, postdocs, faculty and staff. It is not a formal course, but can optionally be taken as a project course, for which you will receive course credit. In fall 2025 we focused on experimental design.
Contact: enawe [at] dtu.dk