Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.03973 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911487702859776 |
|---|---|
| author | Tiofack, Franki Nguimatsia Hellard, Théotime Le Schramm, Fabian Perrin-Gilbert, Nicolas Carpentier, Justin |
| author_facet | Tiofack, Franki Nguimatsia Hellard, Théotime Le Schramm, Fabian Perrin-Gilbert, Nicolas Carpentier, Justin |
| contents | Offline reinforcement learning often relies on behavior regularization that enforces policies to remain close to the dataset distribution. However, such approaches fail to distinguish between high-value and low-value actions in their regularization components. We introduce Guided Flow Policy (GFP), which couples a multi-step flow-matching policy with a distilled one-step actor. The actor directs the flow policy through weighted behavior cloning to focus on cloning high-value actions from the dataset rather than indiscriminately imitating all state-action pairs. In turn, the flow policy constrains the actor to remain aligned with the dataset's best transitions while maximizing the critic. This mutual guidance enables GFP to achieve state-of-the-art performance across 144 state and pixel-based tasks from the OGBench, Minari, and D4RL benchmarks, with substantial gains on suboptimal datasets and challenging tasks. Webpage: https://simple-robotics.github.io/publications/guided-flow-policy/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_03973 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning Tiofack, Franki Nguimatsia Hellard, Théotime Le Schramm, Fabian Perrin-Gilbert, Nicolas Carpentier, Justin Machine Learning Artificial Intelligence Offline reinforcement learning often relies on behavior regularization that enforces policies to remain close to the dataset distribution. However, such approaches fail to distinguish between high-value and low-value actions in their regularization components. We introduce Guided Flow Policy (GFP), which couples a multi-step flow-matching policy with a distilled one-step actor. The actor directs the flow policy through weighted behavior cloning to focus on cloning high-value actions from the dataset rather than indiscriminately imitating all state-action pairs. In turn, the flow policy constrains the actor to remain aligned with the dataset's best transitions while maximizing the critic. This mutual guidance enables GFP to achieve state-of-the-art performance across 144 state and pixel-based tasks from the OGBench, Minari, and D4RL benchmarks, with substantial gains on suboptimal datasets and challenging tasks. Webpage: https://simple-robotics.github.io/publications/guided-flow-policy/ |
| title | Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2512.03973 |