stat.MLNov 17, 2026

Amortizing Tempered Sampling with Corrected Self-Training

Authors: Hoang-Viet Nguyen, Hai-Dang Dau, Alexandre H. Thiéry

Organizations: National University of Singapore · Nanyang Technological University

Abstract

Learning a neural sampler can reduce the cost of repeated sampling from an un-normalized distribution, but obtaining good training data remains difficult. We develop a corrected self-training scheme across a sequence of tempered targets, making use of importance sampling and Markov chain Monte Carlo to correct the samples and replay buffers to store them. We introduce Sequential Score Sampler (SSS) as a particular case. In SSS, the proposal is a score-based diffusion trained by regression on energy gradients; and the samples are corrected by importance weights on the path space. After training, samples can be collected directly from the model, from its importance-corrected proposals, or from the final-temperature buffer. The method amortizes repeated sampling by encoding the global structure of the target in the model. SSS achieves competitive sampling accuracy across the benchmarks and, on several targets, uses shorter runtimes with fewer energy evaluations than competing samplers.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Amortised adaptive importance sampling

    Oct 12, 2026Arran Carter, Víctor Elvira, Esmeralda S. WhitammerImportance SamplingAdaptive Sampling

  2. Adaptive teachers for amortized samplers

    Oct 2, 2024Minsu Kim, Sanghyeok Choi, Taeyoung Yun +7Amortized InferenceRL Exploration

  3. Diffusion Path Samplers via Sequential Monte Carlo

    Jan 29, 2026James Matthew Young, Paula Cordero-Encinar, Sebastian Reich +2Diffusion SamplingSampling from Unnormalized Distributions