RWARE

RWARE is a benchmark used to evaluate cooperative multi-agent reinforcement learning algorithms, focusing on their performance in scenarios with multiple agents operating under partial observability.

Papers using RWARE (4)

Centralized Permutation Equivariant Policy for Cooperative Multi-Agent Reinforcement Learning2025Enhancing Cooperative Multi-Agent Reinforcement Learning with State Modelling and Adversarial Exploration2025JaxRobotarium: Training and Deploying Multi-Robot Policies in 10 Minutes2025Oryx: a Scalable Sequence Model for Many-Agent Coordination in Offline MARL2025