Metraipolis › archive › dc_ec0f404ca1ba

Learning Multimodal One-step Flow Policy via Value-weighted Optimal Transport — arXiv:2609.15883

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.15883
License (as archived)
CC0-1.0
Archived
2026-10-15T22:54:08.185Z
SHA-256 of the archived text
16d5799bb9c65e6662e5de79983f40b3fb37f4a58ea6340d9bbecfb6b4bd02c5
Stable URL of this page
https://metraipolis.com/documento/dc_ec0f404ca1ba

How to cite

Learning Multimodal One-step Flow Policy via Value-weighted Optimal Transport — arXiv:2609.15883 — arXiv (en), https://arxiv.org/abs/2609.15883 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ec0f404ca1ba (Metraipolis archive dc_ec0f404ca1ba, sha256 16d5799bb9c65e6662e5de79983f40b3fb37f4a58ea6340d9bbecfb6b4bd02c5)
BibTeX
@misc{dcec0f404ca1ba,
  title = {Learning Multimodal One-step Flow Policy via Value-weighted Optimal Transport — arXiv:2609.15883},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.15883},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ec0f404ca1ba; sha256 16d5799bb9c65e6662e5de79983f40b3fb37f4a58ea6340d9bbecfb6b4bd02c5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15883. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.15883 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T22:29:03.943Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).