Symbolic regression for defect interactions in 2D materials

TL;DR

This paper explores the use of deep symbolic regression, specifically SEGVAE, to analyze defect interactions in 2D materials, offering interpretable models with results comparable to neural networks.

Contribution

It demonstrates the application of SEGVAE for predicting properties of 2D materials with defects, showing its effectiveness and interpretability compared to graph neural networks.

Findings

SEGVAE achieves comparable results to state-of-the-art GNNs.

Symbolic regression provides interpretable models for material properties.

The method is applicable to natural sciences for data analysis.

Abstract

Machine learning models have become firmly established across all scientific fields. Extracting features from data and making inferences based on them with neural network models often yields high accuracy; however, this approach has several drawbacks. Symbolic regression is a powerful technique for discovering analytical equations that describe data, providing interpretable and generalizable models capable of predicting unseen data. Symbolic regression methods have gained new momentum with the advancement of neural network technologies and offer several advantages, the main one being the interpretability of results. In this work, we examined the application of the deep symbolic regression algorithm SEGVAE to determine the properties of two-dimensional materials with defects. Comparing the results with state-of-the-art graph neural network-based methods shows comparable or, in some cases, even identical outcomes. We also discuss the applicability of this class of methods in natural sciences.