Abstract

Deep Learning has achieved great results on many domains including NLP, speech recognition, computer vision, to name a few. However, it is non-trivial to apply deep learning techniques on graphs. In this talk, Prof. Lai will introduce several important deep learning algorithms developed for graphs, such as deep random walk and graph convolution. Furthermore, Prof. Lai will share his current research on using graph convolutions for heterogenous social networks, and the experimental results on Instagram dataset.

Bio

Prof. Kuan-Ting Lai received bachelor's degree in Electric Engineering and master's degree in Computer Science from National Taiwan University in 2003 and 2005. After graduation, he joined Quanta Computer as a video ASIC engineer, and started to pursue Ph.D. in 2009, under supervision of Prof. Ming-Syan Chen. During 2012-2013, Dr. Lai visited Prof. Shih-Fu Chang’s DVMM lab at Columbia University, and co-developed a large-scale video event detection system with IBM T. J. Watson Research Center. He received his Ph.D. degree in Feb. 2015 and became the VP of technology at Arkados Group. He also co-founded AnyCharge, a wireless charging service provider in Asia. In 2018, Dr. Lai joined the Department of Electronic Engineering at National Taipei University of Technology as an Assistant Professor. His research interests include computer vision, machine learning, deep learning and Internet of Things.

All faculties and students are welcome to join.