Scalable Diffusion for Materials Generation
- URL: http://arxiv.org/abs/2311.09235v2
- Date: Mon, 3 Jun 2024 20:07:00 GMT
- Title: Scalable Diffusion for Materials Generation
- Authors: Sherry Yang, KwangHwan Cho, Amil Merchant, Pieter Abbeel, Dale Schuurmans, Igor Mordatch, Ekin Dogus Cubuk,
- Abstract summary: We develop a unified crystal representation that can represent any crystal structure (UniMat)
UniMat can generate high fidelity crystal structures from larger and more complex chemical systems.
We propose additional metrics for evaluating generative models of materials.
- Score: 99.71001883652211
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Generative models trained on internet-scale data are capable of generating novel and realistic texts, images, and videos. A natural next question is whether these models can advance science, for example by generating novel stable materials. Traditionally, models with explicit structures (e.g., graphs) have been used in modeling structural relationships in scientific data (e.g., atoms and bonds in crystals), but generating structures can be difficult to scale to large and complex systems. Another challenge in generating materials is the mismatch between standard generative modeling metrics and downstream applications. For instance, common metrics such as the reconstruction error do not correlate well with the downstream goal of discovering stable materials. In this work, we tackle the scalability challenge by developing a unified crystal representation that can represent any crystal structure (UniMat), followed by training a diffusion probabilistic model on these UniMat representations. Our empirical results suggest that despite the lack of explicit structure modeling, UniMat can generate high fidelity crystal structures from larger and more complex chemical systems, outperforming previous graph-based approaches under various generative modeling metrics. To better connect the generation quality of materials to downstream applications, such as discovering novel stable materials, we propose additional metrics for evaluating generative models of materials, including per-composition formation energy and stability with respect to convex hulls through decomposition energy from Density Function Theory (DFT). Lastly, we show that conditional generation with UniMat can scale to previously established crystal datasets with up to millions of crystals structures, outperforming random structure search (the current leading method for structure discovery) in discovering new stable materials.
Related papers
- Diffusion Models Are Promising for Ab Initio Structure Solutions from Nanocrystalline Powder Diffraction Data [4.72906860237269]
We present a generative machine learning model based on a Diffusion model that is trained on 45,229 known structures.
We find that our model, PXRDnet, can successfully solve simulated nanocrystals as small as 10 angstroms across 200 materials of varying symmetry and complexity.
arXiv Detail & Related papers (2024-06-16T03:45:03Z) - Generative Inverse Design of Crystal Structures via Diffusion Models with Transformers [1.2289361708127877]
New inorganic materials with promising properties pose a critical challenge, both scientifically and for industrial applications.
Discovery of new inorganic materials with promising properties poses a critical challenge, both scientifically and for industrial applications.
In this study, we explore a new type of diffusion model for the generative inverse design of crystal structures, with a backbone based on a Transformer architecture.
arXiv Detail & Related papers (2024-06-13T16:03:15Z) - Fine-Tuned Language Models Generate Stable Inorganic Materials as Text [57.01994216693825]
Fine-tuning large language models on text-encoded atomistic data is simple to implement yet reliable.
We show that our strongest model can generate materials predicted to be metastable at about twice the rate of CDVAE.
Because of text prompting's inherent flexibility, our models can simultaneously be used for unconditional generation of stable material.
arXiv Detail & Related papers (2024-02-06T20:35:28Z) - Data-Driven Score-Based Models for Generating Stable Structures with
Adaptive Crystal Cells [1.515687944002438]
This work aims at the generation of new crystal structures with desired properties, such as chemical stability and specified chemical composition.
The novelty of the presented approach resides in the fact that the lattice of the crystal cell is not fixed.
A multigraph crystal representation is introduced that respects symmetry constraints, yielding computational advantages.
arXiv Detail & Related papers (2023-10-16T02:53:24Z) - Discovering Interpretable Physical Models using Symbolic Regression and
Discrete Exterior Calculus [55.2480439325792]
We propose a framework that combines Symbolic Regression (SR) and Discrete Exterior Calculus (DEC) for the automated discovery of physical models.
DEC provides building blocks for the discrete analogue of field theories, which are beyond the state-of-the-art applications of SR to physical problems.
We prove the effectiveness of our methodology by re-discovering three models of Continuum Physics from synthetic experimental data.
arXiv Detail & Related papers (2023-10-10T13:23:05Z) - Crystal-GFN: sampling crystals with desirable properties and constraints [103.79058968784163]
We introduce Crystal-GFN, a generative model of crystal structures that sequentially samples structural properties of crystalline materials.
In this paper, we use as objective the formation energy per atom of a crystal structure predicted by a new proxy machine learning model trained on MatBench.
The results demonstrate that Crystal-GFN is able to sample highly diverse crystals with low (median -3.1 eV/atom) predicted formation energy.
arXiv Detail & Related papers (2023-10-07T21:36:55Z) - Evaluating the diversity and utility of materials proposed by generative
models [38.85523285991743]
We show how one state-of-the-art generative model, the physics-guided crystal generation model, can be used as part of the inverse design process.
Our findings suggest how generative models might be improved to enable better inverse design.
arXiv Detail & Related papers (2023-08-09T14:42:08Z) - Unified Model for Crystalline Material Generation [9.940728137241214]
We propose two unified models that act at the same time on crystal lattice and atomic positions.
Our models are capable to learn any arbitrary crystal lattice deformation by lowering the total energy to reach thermodynamic stability.
arXiv Detail & Related papers (2023-06-07T15:23:59Z) - How to See Hidden Patterns in Metamaterials with Interpretable Machine
Learning [82.67551367327634]
We develop a new interpretable, multi-resolution machine learning framework for finding patterns in the unit-cells of materials.
Specifically, we propose two new interpretable representations of metamaterials, called shape-frequency features and unit-cell templates.
arXiv Detail & Related papers (2021-11-10T21:19:02Z) - Learning Neural Generative Dynamics for Molecular Conformation
Generation [89.03173504444415]
We study how to generate molecule conformations (textiti.e., 3D structures) from a molecular graph.
We propose a novel probabilistic framework to generate valid and diverse conformations given a molecular graph.
arXiv Detail & Related papers (2021-02-20T03:17:58Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.