非IIDマルチロボットマニピュレーションのための連合部分空間誘導型視覚-言語-行動ポリシー蒸留
Federated Subspace Guided Vision-Language-Action Policy Distillation for Non-IID Multi-Robot Manipulation
複数ロボットが非IID環境で連合学習する際に、低ランク部分空間と行動分布の蒸留、および互換性に基づくクラスタリングで表現ドリフトを抑え、マニピュレーションポリシーの知識転移を改善する手法を提案。
著者: Biprodip Pal, Kaushik Roy, Yanming Zhu, Brendan Tidd, Alan Wee-Chung Liew, Peyman Moghadam
分類: cs.RO, cs.CV, cs.DC, cs.LG
原文アブストラクト
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.