A Neural-Guided Dynamic Symbolic Network for Exploring Mathematical Expressions from Data
This work addresses scalability and constant-learning issues in symbolic regression for researchers and practitioners in machine learning and scientific discovery, representing an incremental improvement over existing deep generative SR methods.
The authors tackled the problem of symbolic regression (SR) for discovering mathematical expressions from data, proposing DySymNet, a neural-guided dynamic symbolic network that explores symbolic structures via reinforcement learning. The method demonstrated clear superiority over baseline models on standard benchmarks and SRBench with more variables.
Symbolic regression (SR) is a powerful technique for discovering the underlying mathematical expressions from observed data. Inspired by the success of deep learning, recent deep generative SR methods have shown promising results. However, these methods face difficulties in processing high-dimensional problems and learning constants due to the large search space, and they don't scale well to unseen problems. In this work, we propose DySymNet, a novel neural-guided Dynamic Symbolic Network for SR. Instead of searching for expressions within a large search space, we explore symbolic networks with various structures, guided by reinforcement learning, and optimize them to identify expressions that better-fitting the data. Based on extensive numerical experiments on low-dimensional public standard benchmarks and the well-known SRBench with more variables, DySymNet shows clear superiority over several representative baseline models. Open source code is available at https://github.com/AILWQ/DySymNet.