Mahito Sugiyama

I am an associate professor at National Institute of Informatics. My research field is computer science, in particular, I am interested in machine learning, data mining, and knowledge discovery. My central research area is between discrete and continuous worlds.


Our paper "Learning Graph Representation via Formal Concept Analysis" (First author: Yuka Yoneda) has been accepted to NeurIPS 2018 Workshop on Relational Representation Learning (R2L 2018, 2018.12.8) NEW arXiv PDF
Our paper "Bias-Variance Trade-Off in Hierarchical Probabilistic Models Using Higher-Order Feature Interactions" (First author: Simon Luo) has been accepted to The 33rd AAAI Conference on Artificial Intelligence (AAAI 2019, 2019.1.27-2.1) NEW PDF Code (GitHub)
I became a JST PRESTO Researcher (concurrent) NEW
Our paper "Legendre Decomposition for Tensors" has been accepted to The 32nd Annual Conference on Neural Information Processing Systems (NeurIPS 2018, 2018.12.3-8) as spotlight presentation NEW arXiv Paper Poster Slide Code (GitHub)
I have been awarded JSAI Incentive Award
Our paper "graphkernels: R and Python packages for graph comparison" has been accepted to Bioinformatics Paper Library (R) Library (Python)
I give a talk about significant pattern mining on graphs at the 10th International Conference on Multiple Comparison Procedures (MCP 2017, 2017.6.20-23) Slide
Our paper "Tensor Balancing on Statistical Manifold" has been accepted to The 34th International Conference on Machine Learning (ICML 2017, 2017.8.6-11) arXiv PDF Slide Code (GitHub)
I became an associate professor at National Institute of Informatics
A review (in Japanese) about significant pattern mining is published at Communications of the Operations Research Society of Japan. PDF

Old news is here