D4RL

Introduced by Fu et al. in D4RL: Datasets for Deep Data-Driven Reinforcement Learning

D4RL is a collection of environments for offline reinforcement learning. These environments include Maze2D, AntMaze, Adroit, Gym, Flow, FrankKitchen and CARLA.

Source: https://sites.google.com/view/d4rl/home

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
D4RL	D4RL	PMDB
Offline RL	D4RL	KFC
Gym halfcheetah-expert	D4RL	EDAC
Gym halfcheetah-medium-expert	D4RL	SAC-N
Gym halfcheetah-medium-replay	D4RL	SAC-N
Gym halfcheetah-full-replay	D4RL	EDAC
Adroid pen-human	D4RL	EDAC
Adroid hammer-human	D4RL	EDAC
Gym halfcheetah-random	D4RL	EDAC
Adroid relocate-human	D4RL	EDAC
Adroid pen-cloned	D4RL	EDAC
Adroid hammer-cloned	D4RL	EDAC
Adroid door-cloned	D4RL	EDAC
Adroid relocate-cloned	D4RL	SAC-N
Adroid door-human	D4RL	EDAC
Gym halfcheetah-medium	D4RL	SAC-N