Abstract:Sampling from discrete distributions with multiple modes and energy barriers is fundamental to machine learning and computational physics. Recent discrete neural samplers like MDNS suffer from mode collapse and fail to sample high-energy barrier regions between modes, which is critical for free energy estimation and understanding phase transitions. We propose Metadynamics Discrete Neural Sampler (MetaDNS), a general framework integrating well-tempered metadynamics into discrete diffusion or autoregressive samplers. By maintaining an adaptive, history-dependent bias potential along selected low-dimensional coordinates, MetaDNS forces exploration of previously inaccessible regions, enabling free energy reconstruction infeasible with standard neural samplers due to a lack of high-energy samples. On challenging low-temperature benchmarks including Ising, Potts, and the copper-gold binary alloy, MetaDNS reproduces the thermodynamic distribution. Compared to MCMC-based metadynamics, MetaDNS also achieves comparable exploration requiring fewer bias deposition steps.
| Comments: | Accepted at ICML 2026 |
| Subjects: | Statistical Mechanics (cond-mat.stat-mech); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG) |
| ACM classes: | I.2.6; J.2 |
| Cite as: | arXiv:2605.21722 [cond-mat.stat-mech] |
| (or arXiv:2605.21722v1 [cond-mat.stat-mech] for this version) | |
| https://doi.org/10.48550/arXiv.2605.21722 arXiv-issued DOI via DataCite (pending registration) |
From: Xiaochen Du [view email]
[v1]
Wed, 20 May 2026 20:30:15 UTC (4,452 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。