Gym Retro is an open-source platform for reinforcement learning research on games. It has expanded from around 70 Atari and 30 Sega games to over 1,000 games across various emulators. The platform now includes a tool for adding new games. --- Why it matters: This matters because it increases the number of possible environments for researchers to test their reinforcement learning models, potentially leading to more robust and generalizable results. Source: https://openai.com/index/gym-retro