22 lines
647 B
Markdown
22 lines
647 B
Markdown
<img src="./dreamer4-fig2.png" width="400px"></img>
|
|
|
|
## Dreamer 4 (wip)
|
|
|
|
Implementation of Danijar's [latest iteration](https://arxiv.org/abs/2509.24527v1) for his [Dreamer](https://danijar.com/project/dreamer4/) line of work
|
|
|
|
[Temporary Discord](https://discord.gg/MkACrrkrYR)
|
|
|
|
## Citation
|
|
|
|
```bibtex
|
|
@misc{hafner2025trainingagentsinsidescalable,
|
|
title = {Training Agents Inside of Scalable World Models},
|
|
author = {Danijar Hafner and Wilson Yan and Timothy Lillicrap},
|
|
year = {2025},
|
|
eprint = {2509.24527},
|
|
archivePrefix = {arXiv},
|
|
primaryClass = {cs.AI},
|
|
url = {https://arxiv.org/abs/2509.24527},
|
|
}
|
|
```
|