Transition Path Sampling with Improved Off-Policy Training of Diffusion Path Samplers

2024-05-30

AI Summary

Sampling transition pathways between metastable molecular states under high energy barriers remains challenging: conventional molecular dynamics simulations suffer from low efficiency, while AI-based methods relying on hand-crafted collective variables exhibit poor generalizability. To address this, we propose the unbiased and scalable Diffusion Path Sampler (DPS). Our key contributions are: (1) the first learnable control variate framework based on log-variance divergence, integrated with an off-policy reinforcement learning scheme; (2) synergistic incorporation of a replay buffer and simulated annealing to enhance sample efficiency and pathway diversity; and (3) scale-equivariant biased force parameterization, enabling robust modeling of large-scale systems. Evaluated on synthetic potentials, small peptides, and fast-folding proteins, DPS consistently outperforms baselines—generating physically realistic, diverse, and generalizable transition pathways with significantly improved sampling efficiency.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Understanding transition pathways between two meta-stable states of a molecular system is crucial to advance drug discovery and material design. However, unbiased molecular dynamics (MD) simulations are computationally infeasible because of the high energy barriers that separate these states. Although recent machine learning techniques are proposed to sample rare events, they are often limited to simple systems and rely on collective variables (CVs) derived from costly domain expertise. In this paper, we introduce a novel approach that trains diffusion path samplers (DPS) to address the transition path sampling (TPS) problem without requiring CVs. We reformulate the problem as an amortized sampling from the transition path distribution by minimizing the log-variance divergence between the path distribution induced by DPS and the transition path distribution. Based on the log-variance divergence, we propose learnable control variates to reduce the variance of gradient estimators and the off-policy training objective with replay buffers and simulated annealing techniques to improve sample efficiency and diversity. We also propose a scale-based equivariant parameterization of the bias forces to ensure scalability for large systems. We extensively evaluate our approach, termed TPS-DPS, on a synthetic system, small peptide, and challenging fast-folding proteins, demonstrating that it produces more realistic and diverse transition pathways than existing baselines.

Problem

• Molecular Transition Paths
• High Energy Barriers
• Drug and Material Development

Innovation

• Path Sampling
• Molecular Dynamics
• Machine Learning Optimization