Metraipolis › archive › dc_bdedb13eb5ac

Permutation Robustness Is Not Enough: Action Collapse in Multi-Agent Transformer Policies — arXiv:2610.02848

Source
arXiv
Original
https://arxiv.org/abs/2610.02848
License (as archived)
CC0-1.0
Archived
2026-10-05T22:48:05.245Z
SHA-256 of the archived text
70012edee3c8c21698cb99b49144266f9c06f56701794d7ef9279ec08fd1c75b
Stable URL of this page
https://metraipolis.com/documento/dc_bdedb13eb5ac

How to cite

Permutation Robustness Is Not Enough: Action Collapse in Multi-Agent Transformer Policies — arXiv:2610.02848 — arXiv, https://arxiv.org/abs/2610.02848 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bdedb13eb5ac (Metraipolis archive dc_bdedb13eb5ac, sha256 70012edee3c8c21698cb99b49144266f9c06f56701794d7ef9279ec08fd1c75b)
BibTeX
@misc{dcbdedb13eb5ac,
  title = {Permutation Robustness Is Not Enough: Action Collapse in Multi-Agent Transformer Policies — arXiv:2610.02848},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02848},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bdedb13eb5ac; sha256 70012edee3c8c21698cb99b49144266f9c06f56701794d7ef9279ec08fd1c75b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02848. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Transformer policies are attractive for multi-agent robot learning because self-attention can model interactions among agents. However, multi-agent teams are unordered, while transformers typically process agents as ordered token sequences. We study how this mismatch affects cooperative navigation policies under agent-order permutations. Our results show that low permutation error alone can be misleading: policies may appear robust simply because all agents choose the same action. We therefore evaluate policies using both permutation-consistency metrics and action-collapse diagnostics, including action diversity, same-action fraction, and maximum action frequency. A PPO-ID baseline yields non-collapsed behavior but remains order-sensitive, while strong equivariance regularization can still induce homogeneous behavior. A weak equivariance penalty improves the robustness while preserving more diverse actions for teams with \(N=3\) agents, whereas teams with \(N=4\) agents require substantially smaller regularization weights. These findings suggest that multi-agent transformer policies should be evaluated not only by return and permutation robustness, but also by whether they maintain non-collapsed, differentiated multi-agent behavior.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.02848 — 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 2026-10-05T22:45:40.647Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).