Learning the Value Systems of Societies from Preferences
Andrés Holgado-Sánchez, Holger Billhardt, Sascha Ossowski, Sara Degli-Esposti · 2025
Read Learning the Value Systems of Societies from Preferences by Andrés Holgado-Sánchez, Holger Billhardt, Sascha Ossowski, Sara Degli-Esposti (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-20728.txt.
Description
Andrés Holgado-Sánchez, Holger Billhardt, Sascha Ossowski, and Sara Degli-Esposti tackle how to learn the values of a whole society in a 2025 paper. AI systems that act on values need clear models of what people hold dear. Learning one person's values from their choices is hard enough. A society, though, is better seen as many groups with shared but distinct value sets.
The authors define this learning problem in formal terms and propose a method based on deep grouping of data. The method watches the stated choices of a sample of people and learns both shared value grounds and a set of group value systems. They test it on real data about travel choices. The approach could help build AI that respects the mix of values in a community.
Catalog record
| CREATOR | Andrés Holgado-Sánchez, Holger Billhardt, Sascha Ossowski, Sara Degli-Esposti |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-20728