arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.08136cs.LGcs.SYeess.SY
收藏

反向流匹配:基于扩散与流策略的在线强化学习统一框架

Reverse Flow Matching: A Unified Framework for Online Reinforcement Learning with Diffusion and Flow Policies

Zeyang Li(李泽阳);Sunbochen Tang(唐顺波晨);Navid Azizan(阿齐兹安纳维)

Zeyang Li, Sunbochen Tang, Navid Azizan

AI总结:

针对在线强化学习中扩散与流策略缺乏目标样本的问题,提出反向流匹配框架,通过后验均值估计和Langevin Stein算子构造控制变量,统一了噪声期望与梯度期望两类方法,并扩展到流策略,提升训练效率与稳定性。

摘要:

扩散和流策略因其强大的表达能力在在线强化学习(RL)中日益重要,但高效训练它们仍是一个关键挑战。在线RL与标准生成建模的一个根本区别在于缺乏来自Q函数定义的目标玻尔兹曼分布的直接样本。为此,针对扩散策略提出了两类看似不同的方法:噪声期望族,使用噪声的加权平均作为训练目标;梯度期望族,使用Q函数梯度的加权平均。然而,这些目标如何正式相关,或者它们能否被综合成一个更通用的公式,目前尚不清楚。在本文中,我们提出了一个统一框架——反向流匹配(RFM),该框架严格解决了在没有直接目标样本的情况下训练扩散和流模型的问题。通过采用反向推理视角,我们将训练目标表述为给定中间噪声样本的后验均值估计问题。关键地,我们引入Langevin Stein算子来构造零均值控制变量,推导出一类具有相同期望的通用估计器。我们表明,现有的噪声期望和梯度期望方法只是这个更广泛类别中的两个具体实例。这种统一观点带来了两个关键进展:它将针对玻尔兹曼分布的能力从扩散策略扩展到流策略,并使得能够原则性地结合Q值和Q梯度信息形成有效估计器,从而提高训练效率和稳定性。我们将RFM实例化以在在线RL中训练流策略,并在连续控制基准测试中展示了相比扩散策略基线的改进性能。

英文摘要:

Diffusion and flow policies are gaining prominence in online reinforcement learning (RL) due to their expressive power, yet training them efficiently remains a critical challenge. A fundamental difficulty that distinguishes online RL from standard generative modeling is the lack of direct samples from the target Boltzmann distribution defined by the Q-function. To address this, two seemingly distinct families of methods have been proposed for diffusion policies: a noise-expectation family, which uses a weighted average of noise as the training target, and a gradient-expectation family, which employs a weighted average of Q-function gradients. However, it remains unclear how these objectives are formally related, or whether they can be synthesized into a more general formulation. In this paper, we propose a unified framework, reverse flow matching (RFM), which rigorously addresses the problem of training diffusion and flow models without direct target samples. By adopting a reverse inferential perspective, we formulate the training target as a posterior mean estimation problem given an intermediate noisy sample. Crucially, we introduce Langevin Stein operators to construct zero-mean control variates, deriving a general class of estimators that share the same expectation. We show that existing noise-expectation and gradient-expectation methods are simply two specific instances within this broader class. This unified view yields two key advancements: it extends the capability of targeting Boltzmann distributions from diffusion to flow policies, and it enables the principled combination of Q-value and Q-gradient information to form an effective estimator, thereby improving training efficiency and stability. We instantiate RFM to train a flow policy in online RL and demonstrate improved performance on continuous-control benchmarks compared to diffusion policy baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑