MuJoCo Playground

The 'MuJoCo Playground' is a benchmark dataset that contains a variety of simulated environments used to evaluate reinforcement learning algorithms, particularly in the context of robotics and decision-making tasks.

Papers using MuJoCo Playground (4)

Diffusion Policy through Conditional Proximal Policy Optimization2026PolicyFlow: Policy Optimization with Continuous Normalizing Flow in Reinforcement Learning2026Dynamic Sparsity: Challenging Common Sparsity Assumptions for Learning World Models in Robotic Reinforcement Learning Benchmarks2025FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control2025