AI-Newton: A concept-driven physical law discovery system without prior physical knowledge
- 1Peking University
Abstract
While current AI-driven methods excel at deriving empirical models from individual experiments, a significant challenge remains in uncovering the common fundamental physics that underlie these models -- a task at which human physicists are adept. To bridge this gap, we introduce AI-Newton, a novel framework for concept-driven scientific discovery. Our system autonomously derives general physical laws directly from raw, multi-experiment data, operating without supervision or prior physical knowledge. Its core innovations are twofold: (1) proposing interpretable physical concepts to construct laws, and (2) progressively generalizing these laws to broader domains. Applied to a large, noisy dataset of mechanics experiments, AI-Newton successfully rediscovers foundational and universal laws, such as Newton's second law, the conservation of energy, and the universal gravitation. This work represents a significant advance toward autonomous, human-like scientific discovery.
Related
- Large-Language-Model Discovery of Quantum LDPC Codes through Structured Concept Evolution
- From the Light Quantum to the Photon: The Evolution of a Physical Concept
- Data-driven progressive discovery of physical laws
- Physics-Informed and Knowledge-Driven Generative AI for Autonomous Discovery of Porous Oxide Energy Materials: Opportunities and Challenges
- From Autonomous Knowledge Generation to Generalizable Property Prediction: A Precision-First Pipeline for Electro-Optic Materials Discovery
BibTeX
@article{fang2026ainewton,
title = {{AI-Newton: A concept-driven physical law discovery system without prior physical knowledge}},
author = {You-Le Fang and Dong-Shan Jian and Xiang Li and Yan-Qing Ma},
journal = {Physical Review Research},
year = {2026},
month = oct,
doi = {10.1103/dvvc-4rsr},
url = {https://link.aps.org/doi/10.1103/dvvc-4rsr},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - AI-Newton: A concept-driven physical law discovery system without prior physical knowledge
AU - You-Le Fang
AU - Dong-Shan Jian
AU - Xiang Li
AU - Yan-Qing Ma
PY - 2026
DA - 2026/10/09
JO - Physical Review Research
DO - 10.1103/dvvc-4rsr
UR - https://link.aps.org/doi/10.1103/dvvc-4rsr
AB - While current AI-driven methods excel at deriving empirical models from individual experiments, a significant challenge remains in uncovering the common fundamental physics that underlie these models -- a task at which human physicists are adept. To bridge this gap, we introduce AI-Newton, a novel framework for concept-driven scientific discovery. Our system autonomously derives general physical laws directly from raw, multi-experiment data, operating without supervision or prior physical knowledge. Its core innovations are twofold: (1) proposing interpretable physical concepts to construct laws, and (2) progressively generalizing these laws to broader domains. Applied to a large, noisy dataset of mechanics experiments, AI-Newton successfully rediscovers foundational and universal laws, such as Newton's second law, the conservation of energy, and the universal gravitation. This work represents a significant advance toward autonomous, human-like scientific discovery.
SN - 2643-1564
ER -