Mahito Sugiyama

I am an assistant professor at ISIR, Osaka University. 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.


I will give a talk about partial order structure and information geometry at The 19th Information-Based Induction Sciences Workshop (IBIS 2016, 2016.11.16-19) Slide NEW
Our paper "Finding Combinations of Binary Variables with Guaranteed Accuracy" (First author: Yoshito Baba) has been accepted to NIPS 2016 Workshop Adaptive and Scalable Nonparametric Methods in Machine Learning (2016.12.10) PDF NEW
Our paper "Halting in Random Walk Kernels" (Mahito Sugiyama, Karsten Borgwardt) was selected as IEICE TC-IBISML Research Award Finalist
Our paper "Information Decomposition on Structured Space" has been accepted to The 2016 IEEE International Symposium on Information Theory (ISIT 2016, 2016.7.10-15) arXiv Slide Code (GitHub)
Our proposals have been accepted to the Grants-in-Aid for Scientific Research (KAKENHI) "Grant-in-Aid for Young Scientists (B)" (PI) and "Scientific Research (B)" (Co-Investigator)
I gave a talk about statistical data analysis using information geometry at 3rd mini-symposium on Computations, Brains and Machines (2016.3.17) Slide Ref. paper (on arXiv)
I gave a tutorial on Significant Pattern Mining at FIT2015 (2015.9.15-17) Slide
Our paper "Halting in Random Walk Kernels" has been accepted to The 29th Annual Conference on Neural Information Processing Systems (NIPS 2015, 2015.12.7-12) PDF Supplement Poster Code Code (GitHub)

Old news is here