Symbolic regression for defect interactions in 2D materials

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. • Learns human-readable pairwise interaction laws for 2D defects. • Reveals physically adequate kernels. • Predicts formation energy per site and HOMO–LUMO gap from small datasets ( 300 structs). • Competitive with or better than GNNs on key benchmarks while remaining interpretable. • Enables millisecond inverse design via a symbolic surrogate (defect-placement workflow).