Metraipolis › archive › dc_06ee6309ae68

MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning — arXiv:2609.33563

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33563
License (as archived)
CC0-1.0
Archived
2026-10-07T20:54:13.672Z
SHA-256 of the archived text
b3054c9fcbefb79bba5a3a9ff7780fd166db610cc5f6efdcbd56d43f1cc21090
Stable URL of this page
https://metraipolis.com/documento/dc_06ee6309ae68

How to cite

MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning — arXiv:2609.33563 — arXiv (en), https://arxiv.org/abs/2609.33563 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_06ee6309ae68 (Metraipolis archive dc_06ee6309ae68, sha256 b3054c9fcbefb79bba5a3a9ff7780fd166db610cc5f6efdcbd56d43f1cc21090)
BibTeX
@misc{dc06ee6309ae68,
  title = {MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning — arXiv:2609.33563},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33563},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_06ee6309ae68; sha256 b3054c9fcbefb79bba5a3a9ff7780fd166db610cc5f6efdcbd56d43f1cc21090},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33563. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

World models improve sample efficiency by training policies on imagined trajectories, but their usefulness depends on learning representations that capture the information needed for future control. We study whether self-supervised joint-embedding prediction (JEPA) can provide this learning signal for multi-agent reinforcement learning. We introduce MA-JEPA, a stochastic world model that replaces observation reconstruction with prediction of target representations, enabling model-based multi-agent reinforcement learning with centralized training and decentralized execution. A categorical latent state and a causal Transformer are trained with posterior and action-conditioned dynamics prediction objectives and are then used for actor-critic learning from latent imagination. A training-only joint predictor conditions on all agents' local states and actions to predict each agent's next local observation embedding. These predictions are passed through the same local posterior used during real interaction with a centralized critic that is used only for value learning, with execution remaining decentralized. Our experiments show that this architecture performs strongly on SMAC, matching or exceeding the strongest reported comparator mean win rate on four of eight evaluated maps.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.MA).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2609.33563 — 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 2026-10-07T20:24:41.710Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).