About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
AI/Data Informatics: Computational Model Development, Verification, Validation, and Uncertainty Quantification
|
Presentation Title |
Structural Constraint Integration in Generative Model for Discovery of Quantum Material Candidates |
Author(s) |
Ryotaro Okabe, Mouyang Cheng, Abhijatmedhi Chotrattanapituk, Mingda Li |
On-Site Speaker (Planned) |
Ryotaro Okabe |
Abstract Scope |
Recent advancements in machine-learning-based generative models, particularly diffusion models, show great promise for generating new, stable materials. However, challenges remain in integrating geometric patterns into material generation. Here, we introduce Structural Constraint Integration in the GENerative model (SCIGEN). Our system can modify any trained generative diffusion model by strategic masking of the denoised structure with a diffused constrained structure prior to each diffusion step to steer the generation toward constrained outputs. We generate eight million compounds using Archimedean lattices (AL) as prototype constraints, with over 10% surviving a multi-stage stability pre-screening. High-throughput density functional theory (DFT) on 26,000 survived compounds shows that over 50% passed structural optimization at the DFT level. The properties of quantum materials are closely related to the geometric patterns, and thus our results indicate that SCIGEN provides a general framework for generating quantum materials candidates. |
Proceedings Inclusion? |
Planned: |
Keywords |
Computational Materials Science & Engineering, Machine Learning, Magnetic Materials |