The AI Cosmologist I: An Agentic System for Automated Data Analysis
Adam Moss · 2025
Read The AI Cosmologist I: An Agentic System for Automated Data Analysis by Adam Moss (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-2504-03424.txt.
Description
Adam Moss presents an agentic AI system that runs cosmology data analysis on its own, in a 2025 paper. The system covers the full research loop. Specialized agents plan, write code, run experiments, analyze results, and draft reports. Unlike standard auto-tuning tools, it invents varied strategies rather than searching a fixed menu. It also repairs its own execution errors.
Tests spanned several machine learning tasks in the field. The system explored solution spaces and iterated on outcomes. Winning elements from different attempts were combined into new approaches. Example papers produced by the system appear in the appendix. Code and experiment data are public. Moss argues such agents can automate parts of research. That could speed discovery, though human judgment still frames the questions. The work is a first installment in a planned series.
Catalog record
| CREATOR | Adam Moss |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-03424