Saved in:
Bibliographic Details
Main Authors: Lenchner, Jonathan, Srivastava, Karan, Goncalves, Joao, Squillante, Mark, Horesh, Lior
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.00878
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918323288014848
author Lenchner, Jonathan
Srivastava, Karan
Goncalves, Joao
Squillante, Mark
Horesh, Lior
author_facet Lenchner, Jonathan
Srivastava, Karan
Goncalves, Joao
Squillante, Mark
Horesh, Lior
contents Machine-assisted methods for discovering physical laws from background theory and data have recently emerged, promising to advance our understanding of the physical world. However, training and benchmarking these systems remains challenging: real physical theories are limited in number. To address this need, we introduce SynPAT, a system for generating synthetic physical theories with accompanying data. SynPAT produces: (i) a consistent set of axioms forming a synthetic theory, (ii) a symbolic consequence of these axioms representing the discovery target, and (iii) noisy data approximating this consequence. Crucially, to mirror historically incorrect theories (e.g., Newtonian mechanics before Special Relativity), SynPAT can also generate theories whose axioms do not strictly entail, and in fact conflict with, the observed consequence, requiring a correction to the assumed axioms to bridge the gap. We detail SynPAT's methodology and benchmark several open-source symbolic regression systems on our generated theories and data.
format Preprint
id arxiv_https___arxiv_org_abs_2505_00878
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SynPAT: A System for Generating Synthetic Physical Theories with Data
Lenchner, Jonathan
Srivastava, Karan
Goncalves, Joao
Squillante, Mark
Horesh, Lior
Symbolic Computation
I.1.4
Machine-assisted methods for discovering physical laws from background theory and data have recently emerged, promising to advance our understanding of the physical world. However, training and benchmarking these systems remains challenging: real physical theories are limited in number. To address this need, we introduce SynPAT, a system for generating synthetic physical theories with accompanying data. SynPAT produces: (i) a consistent set of axioms forming a synthetic theory, (ii) a symbolic consequence of these axioms representing the discovery target, and (iii) noisy data approximating this consequence. Crucially, to mirror historically incorrect theories (e.g., Newtonian mechanics before Special Relativity), SynPAT can also generate theories whose axioms do not strictly entail, and in fact conflict with, the observed consequence, requiring a correction to the assumed axioms to bridge the gap. We detail SynPAT's methodology and benchmark several open-source symbolic regression systems on our generated theories and data.
title SynPAT: A System for Generating Synthetic Physical Theories with Data
topic Symbolic Computation
I.1.4
url https://arxiv.org/abs/2505.00878