Date: Sep 18, 2020

Time: 12:00 - 13:00

Location: IB 2071

DSRC Seminar | Explaining and Harnessing Adversarial Examples for Robust and Safe Pattern Recognition: Fundamental, Theory and Applications

Adversarial examples are augmented data points generated by imperceptible perturbation of input samples. They have recently drawn much attention with the machine learning community. Being difficult to distinguish from real data, adversarial examples could change the prediction of many state-of-the-art deep learning models. Recent attempts have been made to build robust and safe models that consider adversarial examples. However, these methods can either lead to performance drops, or are ad-hoc in nature and lack mathematic motivations. In this talk, we introduce the fundamental, present the interpretations, and propose a unified framework to build robust machine learning models against adversarial examples in a systematic way. Finally, we also harness adversarial examples and demonstrate a series of successful applications.

Speaker Bio:
Kaizhu Huang is currently a Professor at Xi’an Jiaotong-Liverpool University, China.  He acts as associate dean of research in School of Advanced Technology, XJTLU and is also the founding director of Suzhou Municipal Key Laboratory of Cognitive Computation and Applied Technology. Prof. Huang obtained his PhD degree from Chinese University of Hong Kong (CUHK) in 2004. He worked in Fujitsu Research Centre, CUHK, University of Bristol, National Laboratory of Pattern Recognition, Chinese Academy of Sciences from 2004 to 2012. Prof. Huang has been working in machine learning, neural information processing, and pattern recognition. He was the recipient of 2011 Asia Pacific Neural Network Society Young Researcher Award. He received best paper or book award five times. Until October 2019, he has published 9 books and over 190 international research papers (70+ international journals) e.g., in journals (JMLR, Neural Computation, IEEE T-PAMI, IEEE T-NNLS, IEEE T-BME, IEEE T-Cybernetics) and conferences (NeurIPS, IJCAI, SIGIR, UAI, CIKM, ICDM, ICML, ECML, CVPR). He serves as associated editors/advisory board members in a number of journals and book series. He was invited as keynote speaker in more than 20 international conferences or workshops.