Metraipolis › archive › dc_7078072439fa
Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation Learning — arXiv:2609.33336
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33336
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:01:18.340Z
- SHA-256 of the archived text
8637424440c20efe0adce7904e5971b64648a7d39e34156e50e66f47554c63ec- Stable URL of this page
https://metraipolis.com/documento/dc_7078072439fa
How to cite
Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation Learning — arXiv:2609.33336 — arXiv (en), https://arxiv.org/abs/2609.33336 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7078072439fa (Metraipolis archive dc_7078072439fa, sha256 8637424440c20efe0adce7904e5971b64648a7d39e34156e50e66f47554c63ec)
BibTeX
@misc{dc7078072439fa,
title = {Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation Learning — arXiv:2609.33336},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33336},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7078072439fa; sha256 8637424440c20efe0adce7904e5971b64648a7d39e34156e50e66f47554c63ec},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33336. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Model-based imitation learning (MBIL) improves real-environment interaction efficiency by optimizing policies on imagined rollouts from a learned world model. However, the gap between model-induced and real-environment occupancies makes policy learning sensitive to model error. Conservative MBIL mitigates model exploitation during policy optimization, but when real-environment interactions are collected by the same conservative policy, uncertain regions around the expert distribution remain insufficiently sampled. Generic uncertainty-driven exploration, on the other hand, may allocate interaction to novel but task-irrelevant dynamics. We propose REcoverability-CONditioned Exploration for Model-Based Imitation Learning (RECON). RECON separates conservative policy learning from active data collection by maintaining a main policy for task execution and an explorer for real-environment interaction. The explorer is optimized based on epistemic uncertainty conditioned on recoverability estimated from multi-step main-policy imagination, focusing data collection on unknown states from which the main policy can still return toward expert behavior. Experiments on locomotion, navigation and manipulation show consistent gains in interaction efficiency, imitation performance, and robustness, indicating that RECON directs real-environment interaction toward recovery regions around the expert distribution that are underexplored by prior methods, and thereby learns a world model better suited for imitation. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.33336 — 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 2026-10-07T20:29:02.616Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).