MEIDNet: Multimodal generative AI framework for inverse materials design

TL;DR

MEIDNet is a multimodal generative AI framework that accelerates inverse materials design by integrating structural and property data through equivariant graph neural networks and contrastive learning, enabling efficient discovery of novel materials.

Contribution

The paper introduces MEIDNet, a novel multimodal inverse design framework combining equivariant GNNs and contrastive learning for accelerated materials discovery.

Findings

Achieves 60x higher learning efficiency than traditional methods.

Generates low-bandgap perovskite structures with 13.6% SUN rate.

Demonstrates scalability and adaptability across chemical space.

Abstract

In this work, we present Multimodal Equivariant Inverse Design Network (MEIDNet), a framework that jointly learns structural information and materials properties through contrastive learning, while encoding structures via an equivariant graph neural network (EGNN). By combining generative inverse design with multimodal learning, our approach accelerates the exploration of chemical-structural space and facilitates the discovery of materials that satisfy predefined property targets. MEIDNet exhibits strong latent-space alignment with cosine similarity 0.96 by fusion of three modalities through cross-modal learning. Through implementation of curriculum learning strategies, MEIDNet achieves ~60 times higher learning efficiency than conventional training techniques. The potential of our multimodal approach is demonstrated by generating low-bandgap perovskite structures at a stable, unique, and novel (SUN) rate of 13.6 %, which are further validated by ab initio methods. Our inverse design framework demonstrates both scalability and adaptability, paving the way for the universal learning of chemical space across diverse modalities.