Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
MuZero learns a model of a game to plan moves without being given its rules.
2019
4,346
Works
Selected research papers and books, linked to their authors.
World Models
Trained agents inside a learned simulation of their environment before testing them in the original.
2018
3,121
A Generalist Agent
Gato brought language, games, image descriptions and robot control into a single model.
2022
1,841