Gym Retro
OpenAI Blog
OpenAI released Gym Retro, a full reinforcement learning platform for training AI agents on video games. The platform now supports over 1,000 games across multiple emulators, up from the previous 100 games available. The release includes tools for researchers to add new games to the platform themselves.
Why it matters
We’re releasing the full version of Gym Retro, a platform for reinforcement learning research on games. This brings our publicly-released game count from around 70 Atari games and 30 Sega games to over 1,000 games across a variety of backing emulators. We’re also releasing the tool we use to add new games to the platform.
