Advanced General Intelligence for Science Program Universal Foundation Model for Scientific Research Team
Team Director: Yoshitaka Ushiku (Ph.D.)
Research Summary
Our team develops a common foundation models—including large multimodal models—for Science by AI. We aim to build a multi-AI agent system that collaborates with researchers to autonomously execute research workflows encompassing steps such as literature survey, hypothesis generation, experimental verification, and reporting. While extending this system to diverse disciplines including bio and materials sciences, we seek to realize a foundational system for science capable of trial-and-error through both computational simulation and real-world experimentation.
Main Research Fields
- Informatics
Related Research Fields
- Human informatics and related fields
Keywords
- Natural Language Processing
- Design
- Generative Artificial Intelligence
- Machine Learning
- Multimodal
Selected Publications
Papers with an asterisk(*) are based on research conducted outside of RIKEN.
- 1.
*Tatsunori Taniai, Ryo Igarashi, Yuta Suzuki, Naoya Chiba, Kotaro Saito, Yoshitaka Ushiku, Kanta Ono
"CRYSTALFORMER: INFINITELY CONNECTED ATTENTION FOR PERIODIC STRUCTURE ENCODING"
Published as a conference paper at ICLR 2024 - 2.
*Yoshitomo Matsubara, Naoya Chiba, Ryo Igarashi, Yoshitaka Ushiku
"Rethinking Symbolic Regression Datasets and Benchmarks for Scientific Discovery"
Journal of Data-centric Machine Learning Research (2024) - 3.
*Naoya Chiba, Yuta Suzuki, Tatsunori Taniai, Ryo Igarashi, Yoshitaka Ushiku, Kotaro Saito & Kanta Ono
"Neural structure fields with application to crystal structure autoencoders"
Commun Mater 4, 106 (2023)
Related Links
Lab Members
Principal investigator
- Yoshitaka Ushiku
- Team Director
Contact Information
2-1 Hirosawa, Wako,
Saitama 351-0198, Japan
Email: trip_pr@ml.riken.jp