D-4RL AntMaze

The D-4RL AntMaze is a benchmark dataset used to evaluate offline reinforcement learning algorithms, specifically in the context of sparse-reward, long-horizon tasks involving navigation and manipulation in a maze environment.

Papers using D-4RL AntMaze (2)

Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning2025Model-based Offline Reinforcement Learning with Lower Expectile Q-Learning2024 · 1 cites