[20p-C601-1~16] 23.1 Joint Session N "Informatics"

3:30 PM - 3:45 PM JST(6:30 AM - 6:45 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation AwardHighlighted Presentation

[20p-C601-8] Multimodal Deep Learning for Predicting Diverse Properties pf Functional Materials and Elucidation of Pareto Frontiers

〇Shun Muroga¹, Yasuaki Miki¹, Kenji Hata¹ (1.AIST)

Keywords:

multimodal deep learning,materials informatics,deep learning

As a new data-driven approach applicable to complex materials, we have proposed and demonstrated multimodal deep learning, successfully predicting diverse physical properties. In this talk, we will discuss the strategies for model construction, methods for exploiting Pareto frontiers, and future prospects of multimodal deep learning for a broad discussion on data-driven approaches to materials.