Learning Probabilistic Compositional Representation of Crystalline Materials

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Machine learning (ML) has seen promising developments in materials science, yet its efficacy largely depends on detailed crystal structural data, which are often complex and hard to obtain, limiting their applicability in real-world material synthesis processes. An alternative, using compositional descriptors, offers a simpler approach by indicating the elemental ratios of compounds without detailed structural insights. However, accurately representing materials solely with compositional descriptors presents challenges due to polymorphism, where a single composition can correspond to various structural arrangements, creating ambiguities in its representation. To this end, we introduce PCRL, a novel approach that employs probabilistic modeling of composition to capture the diverse polymorphs from available structural information. Extensive evaluations on sixteen datasets demonstrate the effectiveness of PCRL in learning compositional representation, and analysis on model uncertainty highlights its potential applicability of PCRL in material discovery.
Funding Information
  • Institute of Information & communications Technology Planning & Evaluation (IITP) (RS-2025-02304967)
  • National Research Foundation of Korea (NRF) (RS-2024-00335098)
  • Information & Communications Technology Planning & Evaluation (IITP) (RS-2022-II220157)

This publication has 26 references indexed in Scilit: