Gym Retro
OpenAI just open-sourced 1,000+ games for AI training. Here's why that matters for RL research.
Why it matters
OpenAI expanded Gym Retro from ~100 games to 1,000+, dramatically increasing the dataset available for reinforcement learning research. This infrastructure play lowers the barrier for RL experimentation and accelerates the pace of agent development.
The key facts
10 to knowGame library expanded from ~70 Atari + 30 Sega games to 1,000+ games
Multiple emulator backing support
Open-sourced tooling for adding new games to platform
Publicly released research infrastructure
Published May 25, 2018
1,000+ games across multiple emulators (Atari, Sega, others)
Previously: ~70 Atari + 30 Sega games (~100 total)
Released tooling for adding new games to platform
Publicly available on OpenAI Gym
Reinforcement learning research focus
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: 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…