Newton to Einstein: Axiom-Based Discovery via Game Design
Pingchuan Ma, Benjamin Jones, Tsun-Hsuan Wang, Minghao Guo, Michal Piotr Lipiec, Chuang Gan, Wojciech Matusik · 2025
Read Newton to Einstein: Axiom-Based Discovery via Game Design by Pingchuan Ma, Benjamin Jones, Tsun-Hsuan Wang, Minghao Guo, Michal Piotr Lipiec, Chuang Gan, Wojciech Matusik (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-2509-05448.txt.
Introduction
Scientific discovery by machines should reason from axioms rather than only spot patterns, Pingchuan Ma and six co-authors argue in a 2025 position paper. Ordinary machine learning works inside fixed assumptions, so it struggles to invent genuinely new theory.
Their proposal recasts inquiry as game design: agents act in worlds governed by stated rules and may rewrite those rules when observations do not fit. Early experiments with logic-based games show agents evolving axioms that solve problems previously out of reach. The authors offer the framework as a base for systems that discover in a creative, readable, theory-driven way. The title gestures from Newton's rules toward Einstein's deeper rewrite.
Catalog record
| CREATOR | Pingchuan Ma, Benjamin Jones, Tsun-Hsuan Wang, Minghao Guo, Michal Piotr Lipiec, Chuang Gan, Wojciech Matusik |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 425 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-05448