Genie: Generative Interactive Environments
Abstract
We introduce Genie, the first generative interactive environment trained in an unsupervised manner from unlabelled Internet videos. The model can be prompted to generate an endless variety of action-controllable virtual worlds described through text, synthetic images, photographs, and even sketches. At 11B parameters, Genie can be considered a foundation world model. It is comprised of a spatiotemporal video tokenizer, an autoregressive dynamics model, and a simple and scalable latent action model. Genie enables users to act in the generated environments on a frame-by-frame basis despite training without any ground-truth action labels or other domain specific requirements typically found in the world model literature. Further the resulting learned latent action space facilitates training agents to imitate behaviors from unseen videos, opening the path for training generalist agents of the future.
Cite
Text
Bruce et al. "Genie: Generative Interactive Environments." International Conference on Machine Learning, 2024.Markdown
[Bruce et al. "Genie: Generative Interactive Environments." International Conference on Machine Learning, 2024.](https://mlanthology.org/icml/2024/bruce2024icml-genie/)BibTeX
@inproceedings{bruce2024icml-genie,
title = {{Genie: Generative Interactive Environments}},
author = {Bruce, Jake and Dennis, Michael D and Edwards, Ashley and Parker-Holder, Jack and Shi, Yuge and Hughes, Edward and Lai, Matthew and Mavalankar, Aditi and Steigerwald, Richie and Apps, Chris and Aytar, Yusuf and Bechtle, Sarah Maria Elisabeth and Behbahani, Feryal and Chan, Stephanie C.Y. and Heess, Nicolas and Gonzalez, Lucy and Osindero, Simon and Ozair, Sherjil and Reed, Scott and Zhang, Jingwei and Zolna, Konrad and Clune, Jeff and De Freitas, Nando and Singh, Satinder and Rocktäschel, Tim},
booktitle = {International Conference on Machine Learning},
year = {2024},
pages = {4603-4623},
volume = {235},
url = {https://mlanthology.org/icml/2024/bruce2024icml-genie/}
}