Stoichiometry Representation Learning with Polymorphic Crystal Structures

Published: 28 Oct 2023, Last Modified: 23 Nov 2023NeurIPS2023-AI4Science PosterEveryoneRevisionsBibTeX
Keywords: Materials Science, Stoichiometry, Crystalline Materials, Representation Learning, Graph Neural Networks
Abstract: Despite the recent success of machine learning (ML) in materials science, its success heavily relies on the structural description of crystal, which is itself computationally demanding and occasionally unattainable. Stoichiometry descriptors can be an alternative approach, which reveals the ratio between elements involved to form a certain compound without any structural information. However, it is not trivial to learn the representations of stoichiometry due to the nature of materials science called polymorphism, i.e., a single stoichiometry can exist in multiple structural forms due to the flexibility of atomic arrangements, inducing uncertainties in representation. To this end, we propose PolySRL, which learns the probabilistic representation of stoichiometry by utilizing the readily available structural information, whose uncertainty reveals the polymorphic structures of stoichiometry. Extensive experiments on sixteen datasets demonstrate the superiority of PolySRL, and analysis of uncertainties shed light on the applicability of PolySRL in real-world material discovery. The source code for PolySRL is available at
Submission Track: Original Research
Submission Number: 44