We tackle the problem of sampling from intractable high-dimensional density functions, a fundamental task that often appears in machine learning and statistics. We extend recent sampling-based approaches that leverage controlled stochastic processes to model approximate samples from these target densities. The main drawback of these approaches is that the training objective requires full trajectories to compute, resulting in sluggish credit assignment issues due to use of entire trajectories and a learning signal present only at the terminal time. In this work, we present Diffusion Generative Flow Samplers (DGFS), a sampling-based framework where the learning process can be tractably broken down into short partial trajectory segments, via parameterizing an additional "flow function". Our method takes inspiration from the theory developed for generative flow networks (GFlowNets), allowing us to make use of intermediate learning signals. Through various challenging experiments, we demonstrate that DGFS achieves more accurate estimates of the normalization constant than closely-related prior methods.

翻译：我们针对从难以采样的高维密度函数中进行采样的问题展开研究，这是机器学习和统计学中的一项基础任务。我们扩展了近期基于采样的方法，这些方法利用受控随机过程来建模目标密度的近似样本。现有方法的主要缺陷在于训练目标需要计算完整轨迹，导致因使用整个轨迹而出现的延迟信用分配问题，且仅在终端时刻提供学习信号。本文提出扩散生成流采样器（DGFS），这是一种基于采样的框架，通过参数化额外的"流函数"，可将学习过程可处理地分解为短的部分轨迹片段。我们的方法借鉴了生成流网络（GFlowNets）的理论，从而能够利用中间学习信号。通过多项具有挑战性的实验，我们证明DGFS在归一化常数估计精度上显著优于现有的同类方法。