My name is Akifumi Okuno (奥野彰文), a Technical Staff I of Mathematical Statistics team, RIKEN center for Advanced Intelligence Project (AIP) in Japan. I am also a collaborative researcher of Kyoto university AI research unit. I studied mathematical statistics in my bachelor's (2014, B. Eng., Osaka U.) and master's (2016, M. Eng., Osaka U.) under the supervision of Professor Hidetoshi Shimodaira, and I am currently studying theoretical aspects of statistical machine learning towards doctoral degree.
Research Interest : I am mainly interested in relational learning, especially neural network-based graph embedding (GE) and correlation analysis. I developed universal approximation theorems for the NN-based GE; my recent studies (AISTATS2019a, IJCAI2019) provided theoretically-guaranteed highly expressive GE models. Another study of mine provided a general framework for relational learning (AISTATS2019b), and extended it to hyper-relational learning (arXiv:1908.02573), for further theoretical development. Nowadays, my attention is partially drawn to nonparametric statistics, including nonparametric relational-learning (IBIS2019) and optimal nearest neighbour (arXiv:2002.03054). For details, please see my publications, past presentations, and Google Scholar Research Gate DBLP arXiv.
E-mail : oknakfm [at] gmail.com
Address : Room #111, Research Building 15, Kyoto University (Yoshida-Campus), Yoshida-Honmachi, Sakyo-ku, Kyoto, Japan. Postal code: 606-8317.
Conference Papers (Refereed)
Workshop Paper (Refereed)
Misc. (Non-refereed short papers, in Japanese)
International Conferences and Meetings
Domestic Conferences and Meetings
Other Domestic Seminars and Meetings
Datasets: UCI Machine Learning Repository kaggle Open Datasets Open Data on AWS Awesome Public Datasets List of datasets for ML research (wikipedia) arXiv Times (Datasets) Google Dataset Search MSR Open Data
Time-zone converter: timeanddata.com