Metraipolis › archive › dc_826d37f7ffde

Multi-Agent Coordination via Support-Preserving Distillation — arXiv:2610.10087

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10087
License (as archived)
CC0-1.0
Archived
2026-10-08T11:02:19.910Z
SHA-256 of the archived text
6667e9e6046e4a7908a5ec72f256ad8e1f79043a761cef51b93cf4e694dd42fb
Stable URL of this page
https://metraipolis.com/documento/dc_826d37f7ffde

How to cite

Multi-Agent Coordination via Support-Preserving Distillation — arXiv:2610.10087 — arXiv (en), https://arxiv.org/abs/2610.10087 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_826d37f7ffde (Metraipolis archive dc_826d37f7ffde, sha256 6667e9e6046e4a7908a5ec72f256ad8e1f79043a761cef51b93cf4e694dd42fb)
BibTeX
@misc{dc826d37f7ffde,
  title = {Multi-Agent Coordination via Support-Preserving Distillation — arXiv:2610.10087},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10087},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_826d37f7ffde; sha256 6667e9e6046e4a7908a5ec72f256ad8e1f79043a761cef51b93cf4e694dd42fb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10087. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Offline MARL increasingly relies on generative policies to model multimodal joint behavior, typically by distilling a centralized teacher into decentralized one-step actors under the CTDE. We identify a failure mode at the teacher training stage: standard flow-based teachers pair noise with replay targets independently, so nearby noise samples can be routed toward conflicting coordination modes. The teacher then produces samples between valid modes, and because the distillation loss regresses each local actor onto the conditional mean of the teacher's output given local input, this error is not absorbed but propagated to the student. To remove this teacher-side artifact, we propose Mode-Support Semi-Discrete Optimal Transport (MoSDOT), which summarizes multimodal replay into a finite mode support with prescribed capacities and uses conditional semi-discrete optimal transport to assign each noise sample to a single mode before teacher training. We additionally study a shared-randomness variant that uses a shared noise component at execution to expose the residual gap intrinsic to strict-product execution. On controlled diagnostics and offline MARL benchmarks, MoSDOT improves endpoint quality and routing consistency, particularly on datasets exhibiting multimodal joint behavior.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Multiagent Systems (cs.MA); Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.10087 — 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 2026-10-08T10:09:05.155Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).