发表机构
MBZUAI; Avra; Federal Institute of Ceará（穆罕默德·本·扎耶德人工智能大学; Avra公司; 塞阿拉联邦学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出混合GFlowNets理论框架，涵盖连续与离散索引集合，并新提出分层条件GFlowNets，通过状态空间分解实现高效并行训练，显著提升收敛速度和模式覆盖。
AI 中文摘要
为了实现对离散目标分布的采样，学习一个GFlowNets的集成已成为一种常见方法，相比单一采样器，它能实现更好的状态空间探索和收敛性。然而，这些方法通常会给基础模型带来显著的运行时开销，且其概念联系仍难以捉摸。为解决这一问题，我们首先提出一个通用的理论框架来描述GFlowNets的混合模型，并将其具体化为连续索引（CI）和离散索引（DI）集合。一方面，我们证明CI GFlowNets可以通过随机特征扩展的视角来理解，这能在图结构任务中可证明地提升采样器的表达能力，并通过谱移位减少学习不稳定性。另一方面，我们证明DI GFlowNets涵盖了先前GFlowNet训练方法，并为新提出的分层条件（SC）GFlowNets提供了基础。该方法受马尔可夫链的Doob h-变换启发，根据规定的模函数分解状态空间，并限制每个组件从其中的不同子集采样。重要的是，SC GFlowNets支持集中式和组件级易并行训练，我们证明这两种方式都能显著加速学习收敛和模式覆盖，且不引入任何不可忽略的额外计算。
英文摘要
Learning an ensemble of GFlowNets to sample from a discrete target distribution has become a common approach for achieving better state space exploration and convergence than that of a monolithic sampler. However, these methods often add a substantial runtime overhead to the base model, and their conceptual connection remains elusive. To address this, we first propose a general-purpose theoretical framework for describing a mixture of GFlowNets, which we specialize into continuously (CI) and discretely indexed (DI) collections. On the one hand, we show CI GFlowNets can be interpreted through the lens of a random features expansion, provably boosting the sampler's expressivity in graph-structured tasks and reducing learning instability via spectral shifting. On the other hand, we demonstrate DI GFlowNets encompass prior approaches for GFlowNet training and provide the foundation for the newly proposed Stratum-Conditioned (SC) GFlowNets. This method, which is inspired by the Doob's h-transform of Markov chains, decomposes the state space according to a prescribed modular function and restricts each component to sample from a distinct subset of it. Importantly, SC GFlowNets support centralized and component-wise embarrassingly parallel training, and we show both of them significantly speed up learning convergence and mode coverage without introducing any non-negligible extra computation.
