Crystal Diffusion Variational Autoencoder for Periodic Material Generation

by   Tian Xie, et al.

Generating the periodic structure of stable materials is a long-standing challenge for the material design community. This task is difficult because stable materials only exist in a low-dimensional subspace of all possible periodic arrangements of atoms: 1) the coordinates must lie in the local energy minimum defined by quantum mechanics, and 2) global stability also requires the structure to follow the complex, yet specific bonding preferences between different atom types. Existing methods fail to incorporate these factors and often lack proper invariances. We propose a Crystal Diffusion Variational Autoencoder (CDVAE) that captures the physical inductive bias of material stability. By learning from the data distribution of stable materials, the decoder generates materials in a diffusion process that moves atomic coordinates towards a lower energy state and updates atom types to satisfy bonding preferences between neighbors. Our model also explicitly encodes interactions across periodic boundaries and respects permutation, translation, rotation, and periodic invariances. We significantly outperform past methods in three tasks: 1) reconstructing the input structure, 2) generating valid, diverse, and realistic materials, and 3) generating materials that optimize a specific property. We also provide several standard datasets and evaluation metrics for the broader machine learning community.


page 4

page 6

page 9

page 10

page 11

page 12

page 13

page 16


Towards Symmetry-Aware Generation of Periodic Materials

We consider the problem of generating periodic materials with deep model...

Inverse design of crystals using generalized invertible crystallographic representation

Deep learning has fostered many novel applications in materials informat...

Disentanglement with Hyperspherical Latent Spaces using Diffusion Variational Autoencoders

A disentangled representation of a data set should be capable of recover...

Diffusion probabilistic models enhance variational autoencoder for crystal structure generative modeling

The crystal diffusion variational autoencoder (CDVAE) is a machine learn...

Periodic Graph Transformers for Crystal Material Property Prediction

We consider representation learning on periodic graphs encoding crystal ...

A Crystal-Specific Pre-Training Framework for Crystal Material Property Prediction

Crystal property prediction is a crucial aspect of developing novel mate...

Unified Model for Crystalline Material Generation

One of the greatest challenges facing our society is the discovery of ne...

Please sign up or login with your details

Forgot password? Click here to reset