Deep Symbolic Optimization: Reinforcement Learning for Symbolic Mathematics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hayes, Conor F., Da Silva, Felipe Leno, Yang, Jiachen, Mundhenk, T. Nathan, Lee, Chak Shing, Pettit, Jacob F., Santiago, Claudio, Kim, Sookyung, Kim, Joanne T., Solis, Ignacio Aravena, Glatt, Ruben, Goncalves, Andre R., Ladd, Alexander, Solak, Ahmet Can, Desautels, Thomas, Faissol, Daniel, Petersen, Brenden K., Landajuela, Mikel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910947332849664
author Hayes, Conor F.
Da Silva, Felipe Leno
Yang, Jiachen
Mundhenk, T. Nathan
Lee, Chak Shing
Pettit, Jacob F.
Santiago, Claudio
Kim, Sookyung
Kim, Joanne T.
Solis, Ignacio Aravena
Glatt, Ruben
Goncalves, Andre R.
Ladd, Alexander
Solak, Ahmet Can
Desautels, Thomas
Faissol, Daniel
Petersen, Brenden K.
Landajuela, Mikel
author_facet Hayes, Conor F.
Da Silva, Felipe Leno
Yang, Jiachen
Mundhenk, T. Nathan
Lee, Chak Shing
Pettit, Jacob F.
Santiago, Claudio
Kim, Sookyung
Kim, Joanne T.
Solis, Ignacio Aravena
Glatt, Ruben
Goncalves, Andre R.
Ladd, Alexander
Solak, Ahmet Can
Desautels, Thomas
Faissol, Daniel
Petersen, Brenden K.
Landajuela, Mikel
contents Deep Symbolic Optimization (DSO) is a novel computational framework that enables symbolic optimization for scientific discovery, particularly in applications involving the search for intricate symbolic structures. One notable example is equation discovery, which aims to automatically derive mathematical models expressed in symbolic form. In DSO, the discovery process is formulated as a sequential decision-making task. A generative neural network learns a probabilistic model over a vast space of candidate symbolic expressions, while reinforcement learning strategies guide the search toward the most promising regions. This approach integrates gradient-based optimization with evolutionary and local search techniques, and it incorporates in-situ constraints, domain-specific priors, and advanced policy optimization methods. The result is a robust framework capable of efficiently exploring extensive search spaces to identify interpretable and physically meaningful models. Extensive evaluations on benchmark problems have demonstrated that DSO achieves state-of-the-art performance in both accuracy and interpretability. In this chapter, we provide a comprehensive overview of the DSO framework and illustrate its transformative potential for automating symbolic optimization in scientific discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2505_10762
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep Symbolic Optimization: Reinforcement Learning for Symbolic Mathematics
Hayes, Conor F.
Da Silva, Felipe Leno
Yang, Jiachen
Mundhenk, T. Nathan
Lee, Chak Shing
Pettit, Jacob F.
Santiago, Claudio
Kim, Sookyung
Kim, Joanne T.
Solis, Ignacio Aravena
Glatt, Ruben
Goncalves, Andre R.
Ladd, Alexander
Solak, Ahmet Can
Desautels, Thomas
Faissol, Daniel
Petersen, Brenden K.
Landajuela, Mikel
Machine Learning
Neural and Evolutionary Computing
Symbolic Computation
Deep Symbolic Optimization (DSO) is a novel computational framework that enables symbolic optimization for scientific discovery, particularly in applications involving the search for intricate symbolic structures. One notable example is equation discovery, which aims to automatically derive mathematical models expressed in symbolic form. In DSO, the discovery process is formulated as a sequential decision-making task. A generative neural network learns a probabilistic model over a vast space of candidate symbolic expressions, while reinforcement learning strategies guide the search toward the most promising regions. This approach integrates gradient-based optimization with evolutionary and local search techniques, and it incorporates in-situ constraints, domain-specific priors, and advanced policy optimization methods. The result is a robust framework capable of efficiently exploring extensive search spaces to identify interpretable and physically meaningful models. Extensive evaluations on benchmark problems have demonstrated that DSO achieves state-of-the-art performance in both accuracy and interpretability. In this chapter, we provide a comprehensive overview of the DSO framework and illustrate its transformative potential for automating symbolic optimization in scientific discovery.
title Deep Symbolic Optimization: Reinforcement Learning for Symbolic Mathematics
topic Machine Learning
Neural and Evolutionary Computing
Symbolic Computation
url https://arxiv.org/abs/2505.10762