# Princeton and Berkeley researchers scale empowerment learning

> The study uses skill learning to increase agents’ options in complex simulated environments.

- Canonical URL: https://newsverum.com/en/news/princeton-berkeley-study-scales-empowerment-learning-agents
- Publisher: Newsverum
- Section: Technology
- Country: United States
- Published: 2026-10-12T23:33:38+00:00
- Updated: 2026-10-12T23:33:38+00:00
- Event / data period: 2026-10-09
- Other editions: [Türkçe](https://newsverum.com/tr/news/princeton-berkeley-study-scales-empowerment-learning-agents.md), [العربية](https://newsverum.com/ar/news/princeton-berkeley-study-scales-empowerment-learning-agents.md), [Français](https://newsverum.com/fr/news/princeton-berkeley-study-scales-empowerment-learning-agents.md), [Español](https://newsverum.com/es/news/princeton-berkeley-study-scales-empowerment-learning-agents.md), [Deutsch](https://newsverum.com/de/news/princeton-berkeley-study-scales-empowerment-learning-agents.md), [Português](https://newsverum.com/pt/news/princeton-berkeley-study-scales-empowerment-learning-agents.md)

Researchers at Princeton University and the University of California, Berkeley published a method on 9 October for studying agent empowerment at larger scales. The approach estimates how many executable skills are available from a state, then trains a policy to reach states that offer more options.

In Craftax-Classic, the authors report that agents achieved seven to eight of 22 achievements per rollout after one billion environment steps. The paper also describes behaviors that emerged in simulation, such as building shelter. These results concern specific environments and metrics; the authors do not demonstrate general autonomy or performance in deployed robots or systems.

## Data

### Study scope

Training budget in the Craftax-Classic experiment; not a real-world measure

Methodology: Count reported by the paper for its experimental scope; not a national statistic or field result.

| Indicator | Experimental scope count | Preprint publication date |
|---|---|---|
| Training environment steps | 1000000000 | 2026-10-09 |

Sources: <https://arxiv.org/html/2610.13002v1> · Download data: [JSON](https://newsverum.com/api/v1/articles/princeton-berkeley-study-scales-empowerment-learning-agents/data/study-experimental-scope?lang=en&format=json) · [CSV](https://newsverum.com/api/v1/articles/princeton-berkeley-study-scales-empowerment-learning-agents/data/study-experimental-scope?lang=en&format=csv)

## Sources

- [arXiv preprint by Princeton University and UC Berkeley researchers](<https://arxiv.org/html/2610.13002v1>) (Source checked: 2026-10-13; Published: 2026-10-09)

---

Cite the canonical URL above. [Editorial standards and corrections](https://newsverum.com/en/editorial)