Boltzmann Generators -- Sampling Equilibrium States of Many-Body Systems with Deep Learning

TL;DR

Boltzmann Generators leverage deep learning to efficiently generate unbiased equilibrium samples of many-body systems, significantly reducing computational effort in simulating complex condensed matter and protein systems.

Contribution

This work introduces Boltzmann Generators, a novel neural network-based method for one-shot sampling of equilibrium states in many-body systems, bypassing traditional slow simulation techniques.

Findings

Successfully generated unbiased equilibrium samples for condensed matter systems.

Accurately computed free energy differences.

Discovered new configurations without prior reaction coordinate knowledge.

Abstract

Computing equilibrium states in condensed-matter many-body systems, such as solvated proteins, is a long-standing challenge. Lacking methods for generating statistically independent equilibrium samples in "one shot", vast computational effort is invested for simulating these system in small steps, e.g., using Molecular Dynamics. Combining deep learning and statistical mechanics, we here develop Boltzmann Generators, that are shown to generate unbiased one-shot equilibrium samples of representative condensed matter systems and proteins. Boltzmann Generators use neural networks to learn a coordinate transformation of the complex configurational equilibrium distribution to a distribution that can be easily sampled. Accurate computation of free energy differences and discovery of new configurations are demonstrated, providing a statistical mechanics tool that can avoid rare events during sampling without prior knowledge of reaction coordinates.