Mathematical Modeling and Deep Learning for Small-Data AI

• Overview

  Current technologies of artificial intelligence (AI), including deep learning, are extremely useful as long as a massive amount of high-quality data are available for the target task. In contrast, there are many types of small data in the real world such as pandemic disease spreading over social networks, biomedical information like EEG and fMRI, and sensor signals collected by Internet-of-Things (IoT) devices. Discovering knowledge and extracting useful information from such small data are highly important, but many of the technologies for big data are not suitable for such situations. In this project, we study small-data AI by combining mathematical modeling and deep learning from signal processing and machine learning perspectives.

  

Members

SHIMIZU Akinobu  (Institute of Engineering / Professor)
TANAKA Toshihisa  (Institute of Engineering / Professor)
KONDO Toshiyuki  (Institute of Engineering / Professor)
ISHIDA Hiroshi  (Institute of Engineering / Professor)
HAYAKAWA Ryo  (Institute of Engineering / Associate Professor)