Symbolic Regression with Multimodal Large Language Models and Kolmogorov Arnold Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Harvey, Thomas R., Ruehle, Fabian, Fraser-Taliente, Kit, Halverson, James
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908370113396736
author Harvey, Thomas R.
Ruehle, Fabian
Fraser-Taliente, Kit
Halverson, James
author_facet Harvey, Thomas R.
Ruehle, Fabian
Fraser-Taliente, Kit
Halverson, James
contents We present a novel approach to symbolic regression using vision-capable large language models (LLMs) and the ideas behind Google DeepMind's Funsearch. The LLM is given a plot of a univariate function and tasked with proposing an ansatz for that function. The free parameters of the ansatz are fitted using standard numerical optimisers, and a collection of such ansätze make up the population of a genetic algorithm. Unlike other symbolic regression techniques, our method does not require the specification of a set of functions to be used in regression, but with appropriate prompt engineering, we can arbitrarily condition the generative step. By using Kolmogorov Arnold Networks (KANs), we demonstrate that ``univariate is all you need'' for symbolic regression, and extend this method to multivariate functions by learning the univariate function on each edge of a trained KAN. The combined expression is then simplified by further processing with a language model.
format Preprint
id arxiv_https___arxiv_org_abs_2505_07956
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Symbolic Regression with Multimodal Large Language Models and Kolmogorov Arnold Networks
Harvey, Thomas R.
Ruehle, Fabian
Fraser-Taliente, Kit
Halverson, James
Machine Learning
Neural and Evolutionary Computing
Symbolic Computation
We present a novel approach to symbolic regression using vision-capable large language models (LLMs) and the ideas behind Google DeepMind's Funsearch. The LLM is given a plot of a univariate function and tasked with proposing an ansatz for that function. The free parameters of the ansatz are fitted using standard numerical optimisers, and a collection of such ansätze make up the population of a genetic algorithm. Unlike other symbolic regression techniques, our method does not require the specification of a set of functions to be used in regression, but with appropriate prompt engineering, we can arbitrarily condition the generative step. By using Kolmogorov Arnold Networks (KANs), we demonstrate that ``univariate is all you need'' for symbolic regression, and extend this method to multivariate functions by learning the univariate function on each edge of a trained KAN. The combined expression is then simplified by further processing with a language model.
title Symbolic Regression with Multimodal Large Language Models and Kolmogorov Arnold Networks
topic Machine Learning
Neural and Evolutionary Computing
Symbolic Computation
url https://arxiv.org/abs/2505.07956