D-4RL MuJoCo tasks
Emerging5papers using it
2024first seen
The 'D4RL MuJoCo tasks' is a benchmark dataset used to evaluate the performance of offline reinforcement learning algorithms by providing a variety of static datasets that include expert, suboptimal, and random trajectories in simulated environments.
Papers using D-4RL MuJoCo tasks (5)
- LRT-Diffusion: Calibrated Risk-Aware Guidance for Diffusion PoliciesSymmetric Behavior Regularized Policy OptimizationBayes Adaptive Monte Carlo Tree Search for Offline Model-based Reinforcement LearningExpert or not? assessing data quality in offline reinforcement learningRe:Frame -- Retrieving Experience From Associative Memory