NSF East Asia and Pacific Summer Institute (EAPSI) for FY 2013 in Japan

Root Jonathan C, Boston MA

Investigators

Abstract

This action funds Jonathan Root of Boston University to conduct a research project in the Mathematical and Physical Sciences area during the summer of 2013 at the Institute of Statistical Mathematics in Tachikawa, Tokyo. The project title is High Dimensional Pattern Analysis Algorithms Which Can Operate on General Types of Data. The host scientist is Dr. Kenji Fukumizu. The problem of understanding intelligence, both biological and artificial, is one of the most important scientific questions today. A current approach to this problem is through the topic of learning, or learning from data. The Fellow's research is in kernel methods with applications to learning theory. More specifically the work investigates a large dimensional theory for kernel methods, which has had little theoretical attention. In studying high dimensional kernel methods, it is reasonable to consider a series of domains of diverging dimensionality, for which a series of kernels is introduced. Thus this project considers the behavior of the kernel in the limit as both the dimensionality of X and the sample size diverge. In doing so, the spectrum associated to a kernel is studied. Broader impacts of an EAPSI fellowship include providing the Fellow a first-hand research experience outside the U.S.; an introduction to the science, science policy, and scientific infrastructure of the respective location; and an orientation to the society, culture and language. These activities meet the NSF goal to educate for international collaborations early in the career of its scientists, engineers, and educators, thus ensuring a globally aware U.S. scientific workforce. In addition, the Fellow plans to publish the results of the reseach in peer-reviewed journals. Moreover, the Fellow will speak on the research at seminars at Boston University, and other universities upon invitation.