KST-2023

Abstract

This talk gives an overview of progress and challenges in learnable image encryption with a secret key for deep neural networks (DNNs). Learnable encryption with a secret key enables us not only to protect visual information on plain images but also to embed unique features controlled with a key into images and models. Various applications of such encryption have been developed by using these properties. In this talk, we first focus privacy-preserving image classification tasks with learnable encryption, and then such encryption is demonstrated to give a new insight to adversarially robust defenses and model protection. Finally, we discuss future prospects for reliable deep leaning.

Biography

- Hitoshi Kiya is a Professor of the Department of Computer Science at Tokyo Metropolitan University, Japan. He received the B.E. and M.E. degrees from the Nagaoka University of Technology, Japan, in 1980 and 1982, respectively, and the Dr.Eng. degree from Tokyo Metropolitan University in 1987. In 1982, he joined Tokyo Metropolitan University, where he became a Full Professor in 2000. From 1995 to 1996, he attended the University of Sydney, Australia, as a Visiting Fellow. He is a Life Fellow of IEEE, and a Fellow of IEICE, ITE and AAIA. He served as the President of APSIPA, and the Regional Director-at-Large for Region 10 of the IEEE Signal Processing Society. He was also the President of the IEICE Engineering Sciences Society. He has organized a lot of international conferences in such roles as the TPC Chair of IEEE ICASSP 2012 and as the General Co-Chair of IEEE ISCAS 2019. He has received numerous awards, including 12 best paper awards.