OpenAI Gym

Greg Brockman; Vicki Cheung; Ludwig Pettersson; Jonas Schneider; John Schulman; Jie Tang; Wojciech Zaremba

OpenAI Gym

About

OpenAI Gym is a toolkit for reinforcement learning research. It includes a growing collection of benchmark problems that expose a common interface, and a website where people can share their results and compare the performance of algorithms. This whitepaper discusses the components of OpenAI Gym and the design decisions that went into the software.

Greg Brockman, Vicki Cheung, Ludwig Pettersson, Jonas Schneider, John Schulman, Jie Tang, Wojciech Zaremba · 2016

Related benchmarks

Task	Dataset	Result
Robotic Path Planning	Maze2D (test)	BS1-0.999	22
Reinforcement Learning	Autonomous Driving Dataset	Sample Time (s)31.34	19
Dexterous Manipulation	D4RL Adroit Relocate (offline)	Safety Score0	9
Locomotion Planning	walker2d medium	Safety0.02	7
Locomotion	Hopper Med-Expert	Safety0.05	7
Locomotion Planning	hopper medium	Safety0.05	7
Navigation	Maze2D large	Computation Time (sec)0.02	7
Locomotion	Hopper Med-Expert	Computation Time (sec)0.06	7
Locomotion	hopper medium	Computation Time (sec)0.06	7
Locomotion	walker2d medium	Computation Time (sec)0.06	7