Generative Models for Crystalline Materials

TL;DR

This review explores how generative machine learning models are used to design crystalline materials, focusing on their strengths, limitations, and practical applications.

Contribution

The paper provides a comprehensive analysis of generative models for crystal structure prediction and highlights emerging topics like disorder modeling and synthetic feasibility.

Findings

Generative models are increasingly used for end-to-end crystal structure prediction and de novo generation.

The review evaluates strengths and limitations of various crystal representations and generative approaches.

Emerging topics include modeling defects, synthetic feasibility constraints, and model explainability.

Abstract

Understanding structure‐property relationships in materials is fundamental in condensed matter physics and materials science. Over the past few years, machine learning (ML) has emerged as a powerful tool for advancing this understanding and accelerating materials discovery. Early ML approaches primarily focused on constructing and screening large material spaces to identify promising candidates for various applications. More recently, research efforts have increasingly shifted toward generating crystal structures using end‐to‐end generative models. This review analyzes the current state of generative modeling for crystal structure prediction and de novo generation. It examines crystal representations, outlines the generative models used to design crystal structures, and evaluates their respective strengths and limitations. Furthermore, the review highlights experimental considerations for evaluating generated structures and provides recommendations for suitable existing software tools. Emerging topics, such as modeling disorder and defects, integration in advanced characterization, incorporating synthetic feasibility constraints, and model explainability are explored. Ultimately, this work aims to inform both experimental scientists looking to adapt suitable ML models to their specific circumstances and ML specialists seeking to understand the unique challenges related to inverse materials design and discovery.

Generative machine learning models are increasingly used in crystalline materials design. This review outlines major generative approaches and assesses their strengths and limitations. It also examines how generative models can be adapted to practical applications, discusses key experimental considerations for evaluating generated structures, and highlights emerging directions such as disorder, defects, advanced characterization, synthetic feasibility, and model explainability.

Illustration of end‐to‐end generative machine learning (ML) based inverse materials design for crystalline materials

Figures