OptiFlow：通过价值加权最优传输学习多模态一步流策略

View PDF HTML (experimental)

Abstract:Offline reinforcement learning aims to learn a policy solely from fixed datasets, which often contain multimodal action distributions. Flow policies can naturally represent such multimodal behaviors, but learning an efficient one-step flow policy remains challenging: standard value guidance often leads to mode collapse or exploits overestimation bias in out-of-distribution regions. To address this, we introduce One-step Flow policy via Optimal Transport (OptiFlow), a framework for one-step flow policy learning as a structured sample-allocation problem. OptiFlow jointly trains a value-aware reference flow policy and an efficient one-step policy, coupling their action samples through state-wise entropic optimal transport. For each state, critic-estimated values define the priority of distillation target actions, while the action-distance cost ensures geometrically compatible pairings. By avoiding direct critic maximization, our transport-guided approach enables in-distribution exploitation by anchoring the one-step policy to high-value, dataset-supported modes without the risk of out-of-distribution divergence. Experimental results demonstrate that OptiFlow effectively captures optimal multimodal behaviors and achieves strong performance across diverse offline RL benchmarks. Our code is available at this https URL.

Submission history

From: Jaehun Shon [view email]
[v1] Mon, 14 Sep 2026 17:10:19 UTC (7,314 KB)
[v2] Thu, 29 Oct 2026 12:05:32 UTC (7,314 KB)