Quantum statistics from classical simulations via generative Gibbs sampling

2026-01-28

AI Summary

Efficient simulation of nuclear quantum effects is crucial for molecular modeling, yet conventional path integral molecular dynamics (PIMD) suffers from high computational cost. This work proposes the GG-PI framework, which uniquely integrates generative modeling with Gibbs sampling to recover quantum statistical properties from classical trajectories. Built upon the ring polymer representation, GG-PI employs a generative model to learn the conditional density of a single bead and reconstructs the full quantum distribution via Gibbs sampling. Notably, the method enables transfer across temperatures without retraining. Benchmark tests demonstrate that GG-PI drastically reduces computational overhead, achieving wall-clock times substantially shorter than PIMD while maintaining strong generalization performance.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Sampling/Simulation-based Search
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Accurate simulation of nuclear quantum effects is essential for molecular modeling but expensive using path integral molecular dynamics (PIMD). We present GG-PI, a ring-polymer-based framework that combines generative modeling of the single-bead conditional density with Gibbs sampling to recover quantum statistics from classical simulation data. GG-PI uses inexpensive standard classical simulations or existing data for training and allows transfer across temperatures without retraining. On standard test systems, GG-PI significantly reduces wall clock time compared to PIMD. Our approach extends easily to a wide range of problems with similar Markov structure.

Problem

• nuclear quantum effects
• path integral molecular dynamics
• quantum statistics
• molecular modeling
• classical simulations

Innovation

• generative modeling
• Gibbs sampling
• quantum statistics
• path integral molecular dynamics
• ring polymer