FLAME：面向机器人操作的联邦学习基准

FLAME: A Federated Learning Benchmark for Robotic Manipulation

March 3, 2025

作者: Santiago Bou Betran, Alberta Longhini, Miguel Vasco, Yuchong Zhang, Danica Kragic

cs.AI

摘要

近期机器人操作领域的进展得益于跨多样环境收集的大规模数据集。传统上，这些数据集上的机器人操作策略训练以集中式方式进行，引发了关于可扩展性、适应性和数据隐私的担忧。尽管联邦学习实现了去中心化且保护隐私的训练方式，但其在机器人操作中的应用仍鲜有探索。我们推出了FLAME（跨操作环境的联邦学习），这是首个专为机器人操作中的联邦学习设计的基准。FLAME包含：(i) 一套超过160,000次专家演示的大规模数据集，涵盖多种操作任务，收集自广泛的模拟环境；(ii) 一个在联邦设置下进行机器人策略学习训练与评估的框架。我们在FLAME中评估了标准联邦学习算法，展示了它们在分布式策略学习中的潜力，并指出了关键挑战。该基准为可扩展、自适应且注重隐私的机器人学习奠定了基础。

English

Recent progress in robotic manipulation has been fueled by large-scale datasets collected across diverse environments. Training robotic manipulation policies on these datasets is traditionally performed in a centralized manner, raising concerns regarding scalability, adaptability, and data privacy. While federated learning enables decentralized, privacy-preserving training, its application to robotic manipulation remains largely unexplored. We introduce FLAME (Federated Learning Across Manipulation Environments), the first benchmark designed for federated learning in robotic manipulation. FLAME consists of: (i) a set of large-scale datasets of over 160,000 expert demonstrations of multiple manipulation tasks, collected across a wide range of simulated environments; (ii) a training and evaluation framework for robotic policy learning in a federated setting. We evaluate standard federated learning algorithms in FLAME, showing their potential for distributed policy learning and highlighting key challenges. Our benchmark establishes a foundation for scalable, adaptive, and privacy-aware robotic learning.