Scaling Behavior Cloning Games

Train video game-playing foundation models discovering that increasing training data and network depth enables learning more causal policies. Release 8300+ hours of gameplay data and open-source models for real-time consumer GPU inference.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 809798f432

Frequently asked questions

npx get-skill add adu2021/scaling-behavior-cloning-games