Kernel Methods for Cooperative Multi-Agent Learning with Delays
Cooperative multi-agent decision making involves a group of agents collectively solving individual learning problems, while communicating over a (sparse) network with delays. In this paper, we consider the kernelised contextual bandit problem, where the reward obtained by an agent is an arbitrary linear function of the contexts' images in the related reproducing kernel Hilbert space (RKHS), and a group of agents must cooperate to collectively solve their unique decision problems. We propose Coop-KernelUCB that provides near-optimal bounds on the per-agent regret in this setting, and is both computationally and communicatively efficient. For special cases of the cooperative problem, we also provide variants of Coop-KernelUCB that provides optimal per-agent regret. In addition, our algorithm generalizes several existing results in the multi-agent bandit setting. Finally, on a series of both synthetic and real-world multi-agent network benchmarks, our algorithm significantly outperforms existing clustering or consensus-based algorithms, even in the linear setting.
Code (0)
등록된 구현이 없습니다.
Tasks
ClusteringDecision MakingSimilar Papers 제목 키워드 기반
Kernel Methods for Cooperative Multi-Agent Contextual Bandits
Cooperative multi-agent decision making involves a group of agents cooperatively solving learning problems while communicating over a network with delays. In this paper, we consider the kernelised contextual bandit probl…
Decision MakingMulti-Armed BanditsDelay-Aware Multi-Agent Reinforcement Learning for Cooperative and Competitive Environments
Action and observation delays exist prevalently in the real-world cyber-physical systems which may pose challenges in reinforcement learning design. It is particularly an arduous task when handling multi-agent systems wh…
Autonomous VehiclesDeep Reinforcement LearningMulti-agent Reinforcement Learningreinforcement-learning+2Cooperative Multi-Agent Bandits with Heavy Tails
We study the heavy-tailed stochastic bandit problem in the cooperative multi-agent setting, where a group of agents interact with a common bandit problem, while communicating on a network with delays. Existing algorithms…
Robust Multi-Agent Decision-Making with Heavy-Tailed Payoffs
We study the heavy-tailed stochastic bandit problem in the cooperative multiagent setting, where a group of agents interact with a common bandit problem, while communicating on a network with delays. Existing algorithms …
Decision MakingDACOM: Learning Delay-Aware Communication for Multi-Agent Reinforcement Learning
Communication is supposed to improve multi-agent collaboration and overall performance in cooperative Multi-agent reinforcement learning (MARL). However, such improvements are prevalently limited in practice since most e…
Autonomous DrivingMulti-agent Reinforcement Learningreinforcement-learningReinforcement Learning+1