Inverse Design of Solid-State Materials via a Continuous Representation
By Juhwan Noh, Jae-Hoon Kim, H. Stein, Benjamín Sánchez-Lengeling, J. Gregoire, Alán Aspuru-Guzik, Yousung Jung. Matter, 2019
360 citations recorded by the source corpus.
Summary The non-serendipitous discovery of materials with targeted properties is the ultimate goal of materials research, but to date, materials design lacks the incorporation of all available knowledge to plan the synthesis of the next material. This work presents a framework for learning a continuous representation of materials and building a model for new discovery using latent space representation. The ability of autoencoders to generate experimental materials is demonstrated with vanadium oxides via rediscovery of experimentally known structures when the model was trained without them. Approximately 20,000 hypothetical materials are generated, leading to several completely new metastable VxOy materials that may be synthesizable. Comparison with genetic algorithms suggests computational efficiency of generative models that can explore chemical compositional space effectively by learning the distributions of known materials for crystal structure prediction. These results are an important step toward machine-learned inverse design of inorganic functional materials using generative models.
Subfields: Catalysis, Materials Chemistry
View Paper: https://www.semanticscholar.org/paper/0a7528b6dccd40830b5980f5e02862727af0f81d
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Materials Chemistry
- Tommi S. Jaakkola
- Regina Barzilay
- Octavian-Eugen Ganea
- Tian Xie
- Xiang Fu
- Wengong Jin
- José Miguel Hernández-Lobato
- William L. Hamilton
Citations in the corpus
- Crystal Diffusion Variational Autoencoder for Periodic Material Generation (2021, background): “The first approach treats materials as 3D voxel images, but the process of decoding images back to atom types and coordinates often results in low validity, and the models are not rotationally invariant (Hoffmann et al., 2019; Noh et al., 2019; Court et al., 2020; Long et al., 2021).”