데이터셋 탐색

RLBench is an ambitious large-scale benchmark and learning environment designed to facilitate research in a number of vision-guided manipulation research areas, including: reinforcement learning, imitation learning, multi-task learning, geometric computer vision, and in particular, few-shot learning.

Around 90k different RL runs: 256 hyperparameter configurations for 10 seeds each across a total of 3 algorithms (PPO, SAC, DQN) and 22 environments.