mlsanityData to 7 Oct 2026

Robot Fleet Learning via Policy Merging

arXiv 2310.01362 ↗First posted 2 Oct 2023cs.RO cs.LGcited by 11

Not on the map. Why?

Abstract (arXiv)

Fleets of robots ingest massive amounts of heterogeneous streaming data silos generated by interacting with their environments, far more than what can be stored or transmitted with ease. At the same time, teams of robots should co-acquire diverse skills through their heterogeneous experiences in varied settings. How can we enable such fleet-level learning without having to transmit or centralize fleet-scale data? In this paper, we investigate policy merging (PoMe) from such distributed heterogeneous datasets as a potential solution. To efficiently merge policies in the fleet setting, we propose FLEET-MERGE, an instantiation of distributed learning that accounts for the permutation invariance that arises when parameterizing the control policies with recurrent neural networks. We show that FLEET-MERGE consolidates the behavior of policies trained on 50 tasks in the Meta-World environment, with good performance on nearly all training tasks at test time. Moreover, we introduce a novel robotic tool-use benchmark, FLEET-TOOLS, for fleet policy learning in compositional and contact-rich robot manipulation tasks, to validate the efficacy of FLEET-MERGE on the benchmark.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 61 of its 95 references that are in the map. First posted on arXiv 2 Oct 2023.

Builds on

61 of 95 references.

References · 61

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  4. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
  5. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  6. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
  7. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  8. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,873
  9. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,777
  10. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,530
  11. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  12. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,366
38 more
  1. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  2. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  3. 2015End-to-End Training of Deep Visuomotor Policiescited by 847
  4. 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 791
  5. 2018LEAF: A Benchmark for Federated Settingscited by 780
  6. 2019Linear Mode Connectivity and the Lottery Ticket Hypothesiscited by 627
  7. 2018Federated Learning for Mobile Keyboard Predictioncited by 565
  8. 2020Federated Learning with Matched Averagingcited by 564
  9. 2016Federated Optimization: Distributed Machine Learning for On-Device Intelligencecited by 558
  10. 2022Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulationcited by 545
  11. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learningcited by 517
  12. 2022Git Re-Basin: Merging Models modulo Permutation Symmetriescited by 460
  13. 2018An Algorithmic Perspective on Imitation Learningcited by 455
  14. 2019Bayesian Nonparametric Federated Learning of Neural Networkscited by 385
  15. 2019Which Tasks Should Be Learned Together in Multi-task Learning?cited by 367
  16. 2014In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learningcited by 339
  17. 2020Three Approaches for Personalization with Applications to Federated Learningcited by 321
  18. 2019The Non-IID Data Quagmire of Decentralized Machine Learningcited by 303
  19. 2021The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networkscited by 285
  20. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 226
  21. 2019Model Fusion via Optimal Transportcited by 217
  22. 2018Global Convergence of Policy Gradient Methods for the Linear Quadratic Regulatorcited by 211
  23. 2021On the Origin of Implicit Regularization in Stochastic Gradient Descentcited by 183
  24. 2023ZipIt! Merging Models from Different Tasks without Trainingcited by 178
  25. 2018Learning Latent Permutations with Gumbel-Sinkhorn Networkscited by 172
  26. 2019kPAM: KeyPoint Affordances for Category-Level Robotic Manipulationcited by 152
  27. 2023Equivariant Architectures for Learning in Deep Weight Spacescited by 106
  28. 2020Optimizing Mode Connectivity via Neuron Alignmentcited by 85
  29. 2022Pre-Training for Robots: Offline RL Enables Learning New Tasks from a Handful of Trialscited by 78
  30. 2023Permutation Equivariant Neural Functionalscited by 69
  31. 2019Weight-space symmetry in deep networks gives rise to permutation saddles, connected by equal-loss valleys across the loss landscapecited by 61
  32. 2022Re-basin via implicit Sinkhorn differentiationcited by 61
  33. 2022ToolFlowNet: Robotic Manipulation with Tools via Predicting Tool Flow from Point Cloudscited by 54
  34. 2021kPAM 2.0: Feedback Control for Category-Level Robotic Manipulationcited by 38
  35. 2023Provable Guarantees for Generative Behavior Cloning: Bridging Low-Level Stability and High-Level Behaviorcited by 33
  36. 2022A Unified Framework for Implicit Sinkhorn Differentiationcited by 28
  37. 2020Goal-Auxiliary Actor-Critic for 6D Robotic Grasping with Point Cloudscited by 24
  38. 2021Analysis of the Optimization Landscape of Linear Quadratic Gaussian (LQG) Controlcited by 23

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 1; 2026: 6.

11 papers.

  1. 2024GenSim2: Scaling Robot Data Generation with Multi-modal and Reasoning LLMscited by 48
  2. 2025A Survey on Imitation Learning for Contact-Rich Tasks in Roboticscited by 12
  3. 2026chi0: Resource-Aware Robust Manipulation via Taming Distributional Inconsistenciescited by 11
  4. 2024A survey on secure decentralized optimization and learningcited by 4
  5. 2026Escaping the Diversity Trap in Robotic Manipulation via Anchor-Centric Adaptationcited by 4
  6. 2024Federated Multi-Agent Mapping for Planetary Explorationcited by 2
  7. 2024Combining Federated Learning and Control: A Surveycited by 2
  8. 2026DF-ExpEnse: Diffusion Filtered Exploration for Sample Efficient Finetuningcited by 1
  9. 2026Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shiftscited by 0
  10. 2026An Empirical Study and Open Testbed for Federated Fine-Tuning of Vision-Language-Action Modelscited by 0
  11. 2026Convex AI Compositionality and the Governance of AI System Populationscited by 0