Evolving scientific discovery by unifying data and background knowledge with AI Hilbert

• Published in: Nature communications Vol. 15; no. 1; pp. 5922 - 14
• Main Authors: Cory-Wright, Ryan, Cornelio, Cristina, Dash, Sanjeeb, El Khadir, Bachir, Horesh, Lior
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 14.07.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 639/705/1041; 639/766/259; Axioms; Experimental data; Formulas (mathematics); Gravitational waves; Gravity; Humanities and Social Sciences; Kepler laws; Mixed integer; multidisciplinary; Optimization; Polynomials; Science; Science (multidisciplinary); Scientific laws; Scientific validity; Wave power
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-50074-w

The discovery of scientific formulae that parsimoniously explain natural phenomena and align with existing background theory is a key goal in science. Historically, scientists have derived natural laws by manipulating equations based on existing knowledge, forming new equations, and verifying them experimentally. However, this does not include experimental data within the discovery process, which may be inefficient. We propose a solution to this problem when all axioms and scientific laws are expressible as polynomials and argue our approach is widely applicable. We model notions of minimal complexity using binary variables and logical constraints, solve polynomial optimization problems via mixed-integer linear or semidefinite optimization, and prove the validity of our scientific discoveries in a principled manner using Positivstellensatz certificates. We demonstrate that some famous scientific laws, including Kepler’s Law of Planetary Motion and the Radiated Gravitational Wave Power equation, can be derived in a principled manner from axioms and experimental data. Scientific discovery is a highly relevant task in natural sciences, however generating scientifically meaningful laws and determining their consistency remains challenging. The authors introduce an approach that exploits both experimental data and underlying theory in symbolic form to generate formulas that hold scientific significance by solving polynomial optimization problems.