Foundation Models for Artificial Intelligence
Graph Neural Networks, Recommender Systems, Large Language Models, and High-Performance Computing
Graph Neural Networks, Recommender Systems, Large Language Models, and High-Performance Computing
The Suzumura Laboratory at the University of Tokyo was established in April 2021 and is part of the Department of Information and Communication Engineering within the Graduate School of Information Science and Technology.
Initially, our lab focused on Graph Neural Networks (GNNs), but we are also expanding our research to include various aspects of artificial intelligence and its underlying computing systems, such as large language models and high-performance computing. We recently focus on foundation models for artificial intelligence. Foundation models have emerged as a new paradigm for deep learning models (Bom- masani et al. 2021). FMs are trained on broad data sets and can be used for various downstream tasks. Recently, we have seen the power of FMs in the language domain through ChatGPT/GPT4, Llamas, etc.
We believe it is crucial to explore how these foundational technologies can be applied to a range of real-world applications, including recommender systems, life sciences, sports analytics, and material science. To that end, we have collaborated with various industries, including automotive companies, news organizations, and job-matching services. We welcome new lab members at all levels—master's and Ph.D. students, as well as visiting scholars—and are open to collaborations with both industry and academia worldwide.
Recent News
2024/01 : Our paper titled "Beyond Spatio-Temporal Representations: Evolving Fourier Transform for Temporal Graphs" has been accepted by ICLR 2024 . You can check out the paper here
2023/12: Our paper titled "Optimizing Matching Markets: A Comprehensive Approach Using Graph Neural Networks and Reinforcement Learning" has been accepted by the AAAI EcoSys 2024 Workshop
2023/12: Our paper titled "On the Role of Numerical Encoding in Foundation Model of Sequential Recommendation with Sequential Indexing" has been accepted by the AAAI EcoSys 2024 Workshop
2023/10 : Prof. Toyo Suzumura has started to join the organization committee members of AAAI 2024 (Vancouver, Canada) and serve as a sponsor track chair.
2023/09 : Our AI recommendation paper won the Best Full Paper Runner-Up and the Best Student Paper Awards at ACM RecSys 2023 | Data Science Research Division | Information Technology Center, The University of Tokyo https://www.itc.u-tokyo.ac.jp/academic/en/2023/09/27/post-695/
2023/11: Our following paper has been accepted by the 31st ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2023) (https://sigspatial2023.sigspatial.org/)
Revisiting Mobility Modeling with Graph: A Graph Transformer Model for Next Point-of-Interest Recommendation, Xiaohang Xu (The University of Tokyo), Toyotaro Suzumura (The University of Tokyo), Jiawei Yong (Toyota Motor Corporation), Masatoshi Hanai (The University of Tokyo), Chuang Yang (The University of Tokyo), Hiroki Kanezashi (The University of Tokyo), Renhe Jiang (The University of Tokyo), Shintaro Fukushima (Toyota Motor Corporation)
2021/04 Suzumura Lab has been re-opened at the University of Tokyo