Tue. Sep 8, 2026 9:00 AM - 12:00 PM JST Tue. Sep 8, 2026 12:00 AM - 3:00 AM UTC
N302 (First Year Education Bld. N Block)
Chair : Yuma Iwasaki(NIMS), Masahiro Sato(東大院工)
This session focuses on the development of novel machine learning methods for materials science and their applications. It covers foundational topics such as representation learning, generative AI-based feature design, and improving model interpretability, as well as diverse application cases including catalysis, battery materials, and spectral data analysis, offering a multifaceted discussion of the potential of machine learning.
This session addresses the development of machine learning interatomic potentials (MLIPs) and their applications. In addition to efforts on constructing new potentials and improving their accuracy, it introduces the latest advances in MLIP-driven materials analysis, including molecular dynamics simulations that elucidate local structures, phase transitions, and mechanisms underlying material properties.
This session focuses on materials design and property prediction using generative models and structure search techniques. It broadly introduces data-driven approaches for next-generation materials development, including the generation and prediction of crystal structures, the discovery of novel functional materials, and inverse design for achieving target properties.
