[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125583-en":3,"doc-seo-125583-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},125583,687197100911,"Himbo","https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1785132997149421697",6,"Technology","Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl","PySR3 is an open-source library for practical symbolic regression, aiming to discover human-interpretable symbolic models from scientific data. The work presents a distributed, high-performance system built on SymbolicRegression.jl, featuring a multi-population evolutionary search with an evolve-simplify-optimize loop for optimizing unknown constants in newly found empirical expressions. Runtime operator fusion into SIMD kernels, automatic differentiation, and large-scale cluster distribution are highlighted. The paper also introduces EmpiricalBench to evaluate how well symbolic regression recovers historical empirical equations from original and synthetic datasets.","arXiv :2305 .01582v3 [ astro-ph .IM] 5 May 2023  \nPySR & SymbolicRegression.jl  \ndev  \n[github.com/MilesCranmer/pysr_paper](github.com/MilesCranmer/pysr_paper)  \nInterpretable Machine Learning for Science with PySR and SymbolicRegression.jl  \nMiles Cranmer 1,2  \n1 Princeton University, Princeton, NJ, USA  \n2 Flatiron Institute, New York, NY, USA  \nMay 2, 2023  \nPySR3 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 backend, 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 constantsin newly-discovered empirical expressions. PySR's backend is an extremely optimized Julia library SymbolicRegression.jl4 , 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.  \n1 Introduction  \nJohannes Kepler discovered his famous third law of planetary motion,(period)2 / (radius)3 , from searching for patterns in thirty years of data produced by Tycho Brahe's unaided eye. Kepler did not discover this by searching through all conceivable relationships with a genetic algorithm on a computer, but by his own geometrical intuition.“And it was Kepler's Third Law, not an apple, that led Isaac Newton to discover the law of gravitation” [1] . Likewise, Planck's law was not derived from first principles, but was a symbolic form fit to data [2] . This symbolic relationship would inspire the development of Quantum Mechanics.  \nNow, this first step—discovering empirical rela-  \n[3](3 github.com/MilesCranmer/PySR)[ github.com/MilesCranmer/PySR](3 github.com/MilesCranmer/PySR)  \n[4](4 github.com/MilesCranmer/SymbolicRegression.jl)[ github.com/MilesCranmer/SymbolicRegression.jl](4 github.com/MilesCranmer/SymbolicRegression.jl)  \ntionships from data or based on human intuition—is both difficult and time-consuming, even for lowdimensional data where it has been shown to be NPhard [3] . In modern, high-dimensional datasets, it seems an impossible task to discover simple symbolic relationships without the use of automated tools.  \nThis brings us to an optimization problem known as “symbolic regression,” or SR. In this paper, we describe an algorithm for performing SR, and describe software dedicated to equation discovery in the sciences.  \nSymbolic Regression. SR describes a supervised learning task where the model space is spanned by analytic expressions. This is commonly solved in a multi-objective optimization framework, jointly min-  \nimizing prediction error and model complexity. In this family of algorithms, instead of fitting concrete parameters in some overparameterized general model, one searches the space of simple analytic expressions for accurate and interpretable models. In the history of science, scientists have often performed SR“manually,” using a combination of their intuition and trial-and-error to find simple and accurate empirical expressions. These empirical expressions then might lead to new theoretical insights, such as the aforementioned discoveries of Kepler and Planck leading to classical and quantum mechanics respectively. SR algorithms automate this discovery of empirical equations, exploiting the power of modern computing to test many more expressions than our intuition a","cbCairCkuVbywvr6","https://ap.wps.com/l/cbCairCkuVbywvr6","pdf",4594677,1,24,"English","en",105,"# Introduction\n## Symbolic Regression\n## Equation Discovery for Science\n## PySR3 Overview","[{\"question\":\"What is PySR3 and what problem does it solve?\",\"answer\":\"PySR3 is an open-source library for practical symbolic regression, which discovers human-interpretable symbolic models from data.\"},{\"question\":\"How does PySR’s search algorithm work conceptually?\",\"answer\":\"It uses a multi-population evolutionary strategy with an evolve-simplify-optimize loop designed to optimize unknown scalar constants in newly discovered expressions.\"},{\"question\":\"What capabilities does SymbolicRegression.jl provide for PySR?\",\"answer\":\"SymbolicRegression.jl enables runtime fusion of user-defined operators into SIMD kernels, automatic differentiation, and distribution of expression populations across thousands of cores on a cluster.\"}]","Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl | PDF",1785900046,60,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"interpretable-machine-learning-for-science-with-pysr-and-symbolicregressionjl","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/interpretable-machine-learning-for-science-with-pysr-and-symbolicregressionjl/125583/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is PySR3 and what problem does it solve?","Question",{"text":75,"@type":76},"PySR3 is an open-source library for practical symbolic regression, which discovers human-interpretable symbolic models from data.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does PySR’s search algorithm work conceptually?",{"text":80,"@type":76},"It uses a multi-population evolutionary strategy with an evolve-simplify-optimize loop designed to optimize unknown scalar constants in newly discovered expressions.",{"name":82,"@type":73,"acceptedAnswer":83},"What capabilities does SymbolicRegression.jl provide for PySR?",{"text":84,"@type":76},"SymbolicRegression.jl enables runtime fusion of user-defined operators into SIMD kernels, automatic differentiation, and distribution of expression populations across thousands of cores on a cluster.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,109,112,117,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":29,"slug":108},5,"Comic","comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":110,"slug":111},50,"technology",{"id":113,"doc_module":4,"doc_module_name":46,"category_name":114,"show_sort_weight":115,"slug":116},7,"Healthcare",40,"healthcare",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},8,"Research & Report",30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]