My research is theoretical and centered around the computational principles in the visual cortex of the primate brain. On one hand, I investigate theory of vision to clarify the information processing and representation in the intermediate to higher visual areas regarding complex visual features such as shapes and objects. For this, I use machine learning theories for formalizing statistical learning and inference in the neural visual system and explicitly relate the theories with the experimental data, in collaboration with phyisiology experimentalists. On the other hand, I develop deep learning models to pursue new high-performance artificial intelligence techniques by incorporating discoveries and ideas from neuroscience research. In particular, my recent contributions concentrate on deep generative models that can extract useful information with minimal supervision. I emphasize the importance of such a mutual inspiration between computational neuroscience and the contemporary artificial intelligence research, not only for application but also for cultivation of novel scientific questions.
I have been enjoying since my childhoold playing European classical music with the piano. My recent favorites are French modern pieces, especially, Debussy and Ravel.
Here are some of my performances on YouTube.
I won a few prizes at competitions:
Many people are astonished, but I'm originally from "hard-core" computer science. I used to work on applying discrete mathematics such as automata, logic, and type theory to the design and implementation of novel programming languages in the direction of helping programmers in safety, productivity, and efficiency. My main target in this research direction was to use theory of tree automata to the design of type-safe programming languages specialized to processing XML data. The project ended in 2010.
Foundations of XML Processing: The Tree-automata approach (Cambridge University Press)
"... the process of acceptance will pass through the usual four stages:
--- J.B.S. Haldane (1963)