Gym Retro

OpenAI Blog2018-05-25 15:000 views

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.

Source：https://openai.com/index/gym-retro

My favorites

PreviousOpenAI Fellows Fall 2018 NextAI and compute

What we learned from 48 GitHub lists and 44 pull requests

A GitHub awesome list looks like a straightforward place to share a useful product. Find a relevant section, add a line

0dev.to GitHub

The .claude/settings.local.json gitignore stops working the moment the file is tracked

The Claude Code team's answer to Apiiro's leak research is one sentence: the settings file is meant to stay local, Claud

0dev.to Git

Samestate: one interaction, every device 📱💻

😩 The problem: one site, nine screens, one tab at a time You ship a responsive layout. Then the ritual starts. Open DevT

0dev.to CSS

LibreOffice 26.8.3 fixes 45 issues and becomes more stable

LibreOffice 26.8.3 resolves 45 issues across the 26.8 series, including crashes in Calc, Writer, Impress and Draw and a

0dev.to Linux

My AI agent now tells me when a chat is getting too long

I work with my AI agent all day, and some chats run for hours. At some point the answers get slower, a bit vaguer, and d

0dev.to Python

Introducing Polo: a Wayland MATE compositor

Kerem Söke has announced the 1.0 release of Polo , a Wayland compositor for the MATE Desktop Environment : The idea is t

0LWN.net