Metraipolis › archive › dc_7a1989bd58d5

Bayesian Offline Multi-Agent Reinforcement Learning with Latent World Models — arXiv:2610.39576

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.39576
License (as archived)
CC0-1.0
Archived
2026-10-31T06:06:41.583Z
SHA-256 of the archived text
977822ca3eff86da279b3fbaa25a03e2218d1a5f06ba1e341933b5b61d9164b1
Stable URL of this page
https://metraipolis.com/documento/dc_7a1989bd58d5

How to cite

Bayesian Offline Multi-Agent Reinforcement Learning with Latent World Models — arXiv:2610.39576 — arXiv (en), https://arxiv.org/abs/2610.39576 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7a1989bd58d5 (Metraipolis archive dc_7a1989bd58d5, sha256 977822ca3eff86da279b3fbaa25a03e2218d1a5f06ba1e341933b5b61d9164b1)
BibTeX
@misc{dc7a1989bd58d5,
  title = {Bayesian Offline Multi-Agent Reinforcement Learning with Latent World Models — arXiv:2610.39576},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.39576},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7a1989bd58d5; sha256 977822ca3eff86da279b3fbaa25a03e2218d1a5f06ba1e341933b5b61d9164b1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39576. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Model-based offline multi-agent reinforcement learning (MARL) provides a promising approach to multi-agent decision-making under partial observability, where the global state is unavailable from offline data. However, limited dataset coverage can lead to substantial out-of-distribution (OOD) model errors, which may mislead policy optimization and degrade downstream performance. Existing model-based offline MARL methods typically rely on access to global states or mitigate model errors through pessimistic or conservative policy optimization, leaving the reliability and refinement of learned world models less explored. We propose a Bayesian offline MARL framework that infers latent global states from decentralized observation-action histories and iteratively refines a posterior world model using only offline data. Building on posterior information loss, we extend this criterion to the multi-agent setting as multi-agent posterior information loss (MPIL), which provides a reliability measure for world model refinement and offline evaluation. Theoretically, we establish an upper bound on policy regret that is monotonic in MPIL and further characterize its dependence on the size and state-joint-action coverage of the offline dataset. Experiments on partially observable multi-agent variants of D4RL MuJoCo tasks demonstrate competitive policy performance across diverse offline data conditions, while MPIL-guided refinement improves world model prediction and downstream policy performance. These results support a principled connection between world model reliability, offline data coverage, and policy performance in partially observable offline MARL.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.39576 — 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 2026-10-31T06:06:38.413Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).