Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl

Abstract

PySR is an open-source library for practical symbolic regression, a type of machine learning which aims to discover human-interpretable symbolic models. PySR was developed to democratize and popularize symbolic regression for the sciences, and is built on a high-performance distributed back-end, a flexible search algorithm, and interfaces with several deep learning packages. PySR's internal search algorithm is a multi-population evolutionary algorithm, which consists of a unique evolve-simplify-optimize loop, designed for optimization of unknown scalar constants in newly-discovered empirical expressions. PySR's backend is the extremely optimized Julia library SymbolicRegression.jl, which can be used directly from Julia. It is capable of fusing user-defined operators into SIMD kernels at runtime, performing automatic differentiation, and distributing populations of expressions to thousands of cores across a cluster. In describing this software, we also introduce a new benchmark, "EmpiricalBench," to quantify the applicability of symbolic regression algorithms in science. This benchmark measures recovery of historical empirical equations from original and synthetic datasets.

arXiv comment24 pages, 5 figures, 3 tables. Feedback welcome. Paper source found at https://github.com/MilesCranmer/pysr_paper ; PySR at https://github.com/MilesCranmer/PySR ; SymbolicRegression.jl at https://github.com/MilesCranmer/SymbolicRegression.jl

Cited by in corpus (26)

• KAN-ODEs: Kolmogorov-Arnold Network Ordinary Differential Equations for Learning Dynamical Systems and Hidden Physics
• Discovering interpretable elastoplasticity models via the neural polynomial method enabled symbolic regressions
• A Framework Based on Symbolic Regression Coupled with eXtended Physics-Informed Neural Networks for Gray-Box Learning of Equations of Motion from Data
• Learning Closed-form Equations for Subgrid-scale Closures from High-fidelity Data: Promises and Challenges
• Efficient Generator of Mathematical Expressions for Symbolic Regression
• A precise symbolic emulator of the linear matter power spectrum
• Machine-Learning-Assisted Photonic Device Development: A Multiscale Approach from Theory to Characterization
• TorchSISSO: A PyTorch-Based Implementation of the Sure Independence Screening and Sparsifying Operator for Efficient and Interpretable Model Discovery
• SymbolNet: Neural Symbolic Regression with Adaptive Dynamic Pruning for Compression
• How wavelength affects the hydrodynamic performance of two accelerating mirror-symmetric slender swimmers
• The Inefficiency of Genetic Programming for Symbolic Regression
• Adaptive Uncertainty-Guided Model Selection for Data-Driven PDE Discovery
• cp3-bench: A tool for benchmarking symbolic regression algorithms tested with cosmology
• Improving Genetic Programming for Symbolic Regression with Equality Graphs
• CTSR: Cartesian tensor-based sparse regression for data-driven discovery of high-dimensional invariant governing equations
• rEGGression: an Interactive and Agnostic Tool for the Exploration of Symbolic Regression Models
• Accelerating Graph-based Tracking Tasks with Symbolic Regression
• Numerical Simulation of Three-dimensional High-Lift Configurations Using Data-Driven Turbulence Model
• SINDyG: Sparse Identification of Nonlinear Dynamical Systems from Graph-Structured Data, with Applications to Stuart-Landau Oscillator Networks
• Dimension Reduction for Symbolic Regression
• Neural Infalling Cloud Equations (NICE): Increasing the Efficacy of Subgrid Models and Scientific Equation Discovery using Neural ODEs and Symbolic Regression
• Grammar-based Ordinary Differential Equation Discovery
• The Physics Behind ML-based Quark-Gluon Taggers
• Explainable Hierarchical Deep Learning Neural Networks (Ex-HiDeNN)
• Prompting Neural-Guided Equation Discovery Based on Residuals
• Experimental Design for Missing Physics