PAC Guarantees for Concurrent Reinforcement Learning with Restricted Communication

05/23/2019

∙

We develop model free PAC performance guarantees for multiple concurrent MDPs, extending recent works where a single learner interacts with multiple non-interacting agents in a noise free environment. Our framework allows noisy and resource limited communication between agents, and develops novel PAC guarantees in this extended setting. By allowing communication between the agents themselves, we suggest improved PAC-exploration algorithms that can overcome the communication noise and lead to improved sample complexity bounds. We provide a theoretically motivated algorithm that optimally combines information from the resource limited agents, thereby analyzing the interaction between noise and communication constraints that are ubiquitous in real-world systems. We present empirical results for a simple task that supports our theoretical formulations and improve upon naive information fusion methods.

READ FULL TEXT

PAC Guarantees for Concurrent Reinforcement Learning with Restricted Communication

Communication-Aware Collaborative Learning

Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning

A Hybrid PAC Reinforcement Learning Algorithm

PAC Reinforcement Learning without Real-World Feedback

Crowdsourced PAC Learning under Classification Noise

Ms. Pac-Man Versus Ghost Team CIG 2016 Competition

On Statistical Learning of Simplices: Unmixing Problem Revisited

PAC Guarantees for Concurrent Reinforcement Learning with Restricted Communication

Related Research

Communication-Aware Collaborative Learning

Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning

A Hybrid PAC Reinforcement Learning Algorithm

PAC Reinforcement Learning without Real-World Feedback

Crowdsourced PAC Learning under Classification Noise

Ms. Pac-Man Versus Ghost Team CIG 2016 Competition

On Statistical Learning of Simplices: Unmixing Problem Revisited