Saved in:
Bibliographic Details
Main Authors: Voigt, Henrik, Kahlmeyer, Paul, Lawonn, Kai, Habeck, Michael, Giesen, Joachim
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.19849
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909804077776896
author Voigt, Henrik
Kahlmeyer, Paul
Lawonn, Kai
Habeck, Michael
Giesen, Joachim
author_facet Voigt, Henrik
Kahlmeyer, Paul
Lawonn, Kai
Habeck, Michael
Giesen, Joachim
contents Symbolic regression algorithms search a space of mathematical expressions for formulas that explain given data. Transformer-based models have emerged as a promising, scalable approach shifting the expensive combinatorial search to a large-scale pre-training phase. However, the success of these models is critically dependent on their pre-training data. Their ability to generalize to problems outside of this pre-training distribution remains largely unexplored. In this work, we conduct a systematic empirical study to evaluate the generalization capabilities of pre-trained, transformer-based symbolic regression. We rigorously test performance both within the pre-training distribution and on a series of out-of-distribution challenges for several state of the art approaches. Our findings reveal a significant dichotomy: while pre-trained models perform well in-distribution, the performance consistently degrades in out-of-distribution scenarios. We conclude that this generalization gap is a critical barrier for practitioners, as it severely limits the practical use of pre-trained approaches for real-world applications.
format Preprint
id arxiv_https___arxiv_org_abs_2509_19849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Analyzing Generalization in Pre-Trained Symbolic Regression
Voigt, Henrik
Kahlmeyer, Paul
Lawonn, Kai
Habeck, Michael
Giesen, Joachim
Machine Learning
Artificial Intelligence
Symbolic regression algorithms search a space of mathematical expressions for formulas that explain given data. Transformer-based models have emerged as a promising, scalable approach shifting the expensive combinatorial search to a large-scale pre-training phase. However, the success of these models is critically dependent on their pre-training data. Their ability to generalize to problems outside of this pre-training distribution remains largely unexplored. In this work, we conduct a systematic empirical study to evaluate the generalization capabilities of pre-trained, transformer-based symbolic regression. We rigorously test performance both within the pre-training distribution and on a series of out-of-distribution challenges for several state of the art approaches. Our findings reveal a significant dichotomy: while pre-trained models perform well in-distribution, the performance consistently degrades in out-of-distribution scenarios. We conclude that this generalization gap is a critical barrier for practitioners, as it severely limits the practical use of pre-trained approaches for real-world applications.
title Analyzing Generalization in Pre-Trained Symbolic Regression
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.19849