Deep symbolic regression for physics guided by units constraints: toward the automated discovery of physical laws

Abstract

Symbolic Regression is the study of algorithms that automate the search for analytic expressions that fit data. While recent advances in deep learning have generated renewed interest in such approaches, the development of symbolic regression methods has not been focused on physics, where we have important additional constraints due to the units associated with our data. Here we present $\Phi$-SO, a Physical Symbolic Optimization framework for recovering analytical symbolic expressions from physics data using deep reinforcement learning techniques by learning units constraints. Our system is built, from the ground up, to propose solutions where the physical units are consistent by construction. This is useful not only in eliminating physically impossible solutions, but because the "grammatical" rules of dimensional analysis restrict enormously the freedom of the equation generator, thus vastly improving performance. The algorithm can be used to fit noiseless data, which can be useful for instance when attempting to derive an analytical property of a physical model, and it can also be used to obtain analytical approximations to noisy data. We test our machinery on a standard benchmark of equations from the Feynman Lectures on Physics and other physics textbooks, achieving state-of-the-art performance in the presence of noise (exceeding 0.1%) and show that it is robust even in the presence of substantial (10%) noise. We showcase its abilities on a panel of examples from astrophysics.

arXiv comment29 pages, 9 figures, 11 tables. Accepted for publication at ApJ

References in corpus (22)

• The Structure of Cold Dark Matter Halos
• The Gaia mission
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• The Complete Light-curve Sample of Spectroscopically Confirmed Type Ia Supernovae from Pan-STARRS1 and Cosmological Constraints from The Combined Pantheon Sample
• Interpretable machine learning: definitions, methods, and applications
• SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates
• Science with the Square Kilometer Array: Motivation, Key Science Projects, Standards and Assumptions
• A Bayesian machine scientist to aid in the solution of challenging scientific problems
• AI Poincaré: Machine Learning Conservation Laws from Trajectories
• Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression
• Improving Model-based Genetic Programming for Symbolic Regression of Small Expressions
• A computational framework for physics-informed symbolic regression with straightforward integration of domain knowledge
• Toward an AI Physicist for Unsupervised Learning
• Machine-assisted discovery of relationships in astronomy
• Probabilistic Grammars for Equation Discovery
• Finding universal relations in subhalo properties with artificial intelligence
• Augmenting astrophysical scaling relations with machine learning: application to reducing the Sunyaev-Zeldovich flux-mass scatter
• On the functional form of the radial acceleration relation
• SYMBA: Symbolic Computation of Squared Amplitudes in High Energy Physics with Machine Learning
• The ACTIONFINDER: An unsupervised deep learning algorithm for calculating actions and the acceleration field from a set of orbit segments
• A divide and conquer method for symbolic regression

Cited by in corpus (15)

• Data-Driven Equation Discovery of a Cloud Cover Parameterization
• A precise symbolic emulator of the linear matter power spectrum
• Optimal Inflationary Potentials
• Analysis of strong coupling constant with machine learning and its application
• cp3-bench: A tool for benchmarking symbolic regression algorithms tested with cosmology
• Discovering Nuclear Models from Symbolic Machine Learning
• Approximating neutron-star radii using gravitational-wave only measurements with symbolic regression
• SPINEX_ Symbolic Regression: Similarity-based Symbolic Regression with Explainable Neighbors Exploration
• Dynamics of tidal spiral arms: Machine learning-assisted identification of equations and application to the Milky Way
• Interpretable data-driven turbulence modeling for separated flows using symbolic regression with unit constraints
• Machine Learning Unveils the Power Law of Finite-Volume Energy Shifts
• Comparison of symbolic regression algorithms in Star/galaxy/quasar separation
• UniSymNet: A Unified Symbolic Network Guided by Transformer
• Can AI Dream of Unseen Galaxies? Conditional Diffusion Model for Galaxy Morphology Augmentation
• Segmenting proto-halos with vision transformers