Crystal Diffusion Variational Autoencoder for Periodic Material Generation
This addresses the problem of material design for researchers and engineers by providing a method that generates valid and diverse materials, though it appears incremental as it builds on existing diffusion and VAE approaches.
The paper tackled the challenge of generating stable periodic materials by proposing a Crystal Diffusion Variational Autoencoder (CDVAE) that incorporates physical inductive biases, resulting in significant outperformance in reconstruction, generation, and property optimization tasks compared to past methods.
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.