1. Boltzmann priors for Implicit Transfer Operators
- Author
-
Diez, Juan Viguera, Schreiner, Mathias, Engkvist, Ola, and Olsson, Simon
- Subjects
Physics - Chemical Physics - Abstract
Accurate prediction of thermodynamic properties is essential in drug discovery and materials science. Molecular dynamics (MD) simulations provide a principled approach to this task, yet they typically rely on prohibitively long sequential simulations. Implicit Transfer Operator (ITO) Learning offers a promising approach to address this limitation by enabling stable simulation with time steps orders of magnitude larger than MD. However, to train ITOs, we need extensive, unbiased MD data, limiting the scope of this framework. Here, we introduce Boltzmann Priors for ITO (BoPITO) to enhance ITO learning in two ways. First, BoPITO enables more efficient data generation, and second, it embeds inductive biases for long-term dynamical behavior, simultaneously improving sample efficiency by one order of magnitude and guaranteeing asymptotically unbiased equilibrium statistics. Further, we showcase the use of BoPITO in a new tunable sampling protocol interpolating ITO models trained on off-equilibrium simulation data and an unbiased equilibrium distribution to solve inverse problems in molecular science.
- Published
- 2024