Reinforcement Learning

Atari Games

Playing Atari video games (Atari 2600 benchmark).

1 datasets9 results

Atari Games is a key task in reinforcement learning. Below you will find the standard benchmarks used to evaluate models, along with current state-of-the-art results.

Benchmarks & SOTA

Atari 2600

Arcade Learning Environment (Atari 2600)

20139 results

Suite of 57 Atari 2600 games. Standard benchmark for deep reinforcement learning agents.

State of the Art

Go-Explore

Uber AI

40000

human-normalized-score

Related Tasks

Continuous Control

Control tasks with continuous action spaces (MuJoCo).

Offline RL

Learning from fixed datasets without environment interaction.

Back to Reinforcement Learning