Metraipolis › archive › dc_e014ec9a1741

Federated Subspace Guided Vision-Language-Action Policy Distillation for Non-IID Multi-Robot Manipulation — arXiv:2609.32239

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32239
License (as archived)
CC0-1.0
Archived
2026-10-07T22:12:18.637Z
SHA-256 of the archived text
37ff357f715f169f370c2d7a2766a791959951f9bc7be1adbde46c99727f05ff
Stable URL of this page
https://metraipolis.com/documento/dc_e014ec9a1741

How to cite

Federated Subspace Guided Vision-Language-Action Policy Distillation for Non-IID Multi-Robot Manipulation — arXiv:2609.32239 — arXiv (en), https://arxiv.org/abs/2609.32239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e014ec9a1741 (Metraipolis archive dc_e014ec9a1741, sha256 37ff357f715f169f370c2d7a2766a791959951f9bc7be1adbde46c99727f05ff)
BibTeX
@misc{dce014ec9a1741,
  title = {Federated Subspace Guided Vision-Language-Action Policy Distillation for Non-IID Multi-Robot Manipulation — arXiv:2609.32239},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32239},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e014ec9a1741; sha256 37ff357f715f169f370c2d7a2766a791959951f9bc7be1adbde46c99727f05ff},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Federated learning offers a natural way for multiple robots to jointly improve manipulation policies without requiring centralized access to training demonstrations. However, non-IID task and environment distributions can induce representation drift and mutually incompatible robot-policy updates, making naive parameter aggregation destructive. We present FedDRMan, a federated subspace-guided distillation framework for heterogeneous robot manipulation. At each communication round, the server model provides a frozen teacher for local behavior cloning, while low-rank multimodal subspace and action-distribution distillation preserve globally useful representation geometry and policy behavior. To address heterogeneous aggregation, FedDRMan groups clients by update compatibility and maintains a persistent model for each cluster. The server then spectrally rebalances each compatible aggregate to mitigate attenuation of weaker task-relevant robot-policy update directions. Extensive experiments on LIBERO across diverse non-IID settings, heterogeneity levels, client participation variation, together with ablations and aggregation analyses, show that FedDRMan substantially improves knowledge transfer and consistently outperforms strong federated baselines achieving a peak mean success rate of 80.7%, 11.6 percentage points above the strongest evaluated federated baseline.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-29 in cs.DC).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.32239 — 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 2026-10-07T21:53:13.742Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).