arXiv Machine Learning

Asynchronous Cooperative Online Learning for Multi-Robot Control under Computational Delays

arXiv Machine Learning
Aug 26

Learning to Act While Waiting: RL Finetuning of Generalist Robot Policies Under Inference Latency

The paper introduces ARLI, a latency‑aware framework that enables reinforcement learning fine‑tuning of large generalist robot policies despite inference delays. ARLI combines asynchronous inference with state augmentations—incorporating committed actions and mid‑inference observations—to restore near‑Markovian dynamics and maintain reactivity. Experiments on simulated and real‑world manipulation tasks show that ARLI allows effective policy improvement under latency, outperforming standard RL even in no‑latency scenarios.

By Brian Zhu (Siemens), Momen Khalil (Siemens), E Harrison (UC Berkeley), Emanuele Poggi (Siemens), Philipp Schmitt (Siemens), Bernd Kast (Siemens), Philine Meister (Siemens), Pranav Atreya (UC Berkeley), Qiyang Li (UC Berkeley), Finn Ferchau (Siemens), Cesar Colmenero (Siemens), Yash Shahapurkar (Siemens), Gokul Narayanan (Siemens), Melih Erdogan (Siemens), Kai Wurm (Siemens), Georg von Wichert (Siemens), Oier Mees (Microsoft, ETH Zurich, UC Berkeley), Eugen Solowjow (Siemens), Andrew Wagenmaker (UC Berkeley), Sergey Levine (UC Berkeley)
arXiv Machine Learning
Jun 5

Merging model-based control with multi-agent reinforcement learning for multi-agent cooperative teaming strategies

arXiv:2606. 06011v1 Announce Type: cross Abstract: In this work, we propose a framework that combines multi-agent reinforcement learning (MARL) with model-based control to achieve safe, dynamically feasible actions in cooperative multi-agent tasks.

By Christian Llanes, Spencer W. Jensen, Samuel Coogan
arXiv Machine Learning
1d ago

COIN-GP: Cooperative Online Learning in Networked Distributed Systems with Partial Measurements via Gaussian Process Regression

The paper introduces COIN-GP, an observer‑based cooperative learning framework for distributed sensor networks that jointly estimates system states and partially unknown dynamics using online distributed Gaussian Process regression. It addresses challenges of incomplete state observations and deficient GP models by proposing a novel data collection strategy with theoretical feasibility conditions. The authors also derive an error upper bound for both state and model estimation and demonstrate through simulations that COIN‑GP outperforms existing distributed GP‑based methods.

By Zewen Yang, Xiaobing Dai, Zhenxiao Yin, Hang Zhao, Zhijun Li, C. C. Chan
Hugging Face Trending Papers
Jun 17

Formal Verification of Learned Multi-Agent Communication Policies via Decision Tree Distillation

Multi-agent reinforcement learning (MARL) enables agents to develop coordination strategies through emergent communication, but neural policies lack the formal safety guarantees required for safety-critical robotic deployment in drone swarms and autonomous vehicle fleets. We present the first end-to-end framework for safety verification of learned multi-agent communication policies through policy abstraction: neural policies are distilled into interpretable decision trees, then formally verified, with empirical validation confirming that verified safety properties transfer to original networks.

arXiv AI
Sep 10

D3ARC: Time-Critical Distributed Disaster Detection for Asynchronous Cooperative Multi-Robot Systems

D3ARC is an asynchronous distributed hierarchical framework designed for time‑critical wildfire detection using multiple robotic agents. It enables cooperative perception, shared situational awareness, and coordinated actions while a remote controller asynchronously directs each robot’s motion. The system incorporates safe navigation, coverage efficiency, and a forward‑looking capability to evaluate candidate strategies before execution, achieving up to 94% mission success and 89.4% detection confidence in realistic simulations.

By Nikolaos Koursioumpas, Lina Magoula, Nancy Alonistioti, Ramin Khalili
arXiv AI
Jul 14

Adaptive Reinforcement Learning for Unobservable Random Delays

arXiv:2506. 14411v2 Announce Type: replace-cross Abstract: In standard reinforcement learning (RL) settings, the interaction between the agent and the environment is typically modeled as a Markov decision process (MDP), which assumes that the agent observes the system state instantaneously, selects an action without delay, and executes it immediately.

By John Wikman, Alexandre Proutiere, David Broman