Multi-Agent MuJoCo

'Multi-Agent MuJoCo' is a benchmark that contains various cooperative multi-agent environments used to evaluate the performance and coordination of algorithms in addressing challenges such as non-stationarity and unstable training in multi-agent reinforcement learning.

Papers using Multi-Agent MuJoCo (8)

Bridging MARL to SARL: An Order-Independent Multi-Agent Transformer via Latent Consensus2026Multi-Agent Deep Reinforcement Learning Under Constrained Communications2026Oryx: a Scalable Sequence Model for Many-Agent Coordination in Offline MARL2025Learning Generalizable Skills from Offline Multi-Task Data for Multi-Agent Cooperation2025VDFD: Multi-Agent Value Decomposition Framework with Disentangled World Model2023Optimistic Multi-Agent Policy Gradient2023 · 1 citesHeterogeneous Multi-Agent Reinforcement Learning via Mirror Descent Policy Optimization2023FP3O: Enabling Proximal Policy Optimization in Multi-Agent Cooperation with Parameter-Sharing Versatility2023