Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl

TL;DR

PySR is an open-source symbolic regression library that uses a high-performance distributed algorithm to discover interpretable models in science, introducing a new benchmark for evaluating such algorithms.

Contribution

The paper presents PySR, a scalable, flexible symbolic regression tool with a novel evolve-simplify-optimize algorithm and a new benchmark for scientific applicability.

Findings

PySR efficiently discovers empirical equations from datasets.

The software fuses operators into SIMD kernels for performance.

The EmpiricalBench benchmark assesses symbolic regression in scientific contexts.

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.