【SEMINAR論文研討】110/04/07 Topic:From Machine Learning to Deep Learning: Evolutions and Challenges

Institute of Network Engineering



TOPIC:From Machine Learning to Deep Learning: Evolutions and Challenges

時間:110/04/07 (Wed.) 3:30-5:00PM


講者:黃正能 教授

Prof. Jenq-Neng Hwang

Department of Electrical and  Computer Engineering

University of Washington

Seattle, WA, USA,


Dr. Jenq-Neng Hwang received the BS and MS degrees, both in electrical engineering from the National Taiwan University, Taipei, Taiwan, in 1981 and 1983 separately. He then received his Ph.D. degree from the University of Southern California. In the summer of 1989, Dr. Hwang joined the Department of Electrical and Computer Engineering (ECE) of the University of Washington in Seattle, where he has been promoted to Full Professor since 1999. He served as the Associate Chair for Research from 2003 to 2005, and from 2011-2015. He also served as the Associate Chair for Global Affairs from 2015-2020. He is the founder and co-director of the Information Processing Lab., which has won CVPR AI City Challenges and BMTT Challenge awards in the past years. He has written more than 400 journal, conference papers and book chapters in the areas of machine learning, multimedia signal processing, and multimedia system integration and networking, including an authored textbook on “Multimedia Networking: from Theory to Practice,” published by Cambridge University Press. Dr. Hwang has close working relationship with the industry on multimedia signal processing and multimedia networking. Dr. Hwang received the 1995 IEEE Signal Processing Society's Best Journal Paper Award. He is a founding member of Multimedia Signal Processing Technical Committee of IEEE Signal Processing Society and was the Society's representative to IEEE Neural Network Council from 1996 to 2000. He is currently a member of Multimedia Technical Committee (MMTC) of IEEE Communication Society and also a member of Multimedia Signal Processing Technical Committee (MMSP TC) of IEEE Signal Processing Society. He served as associate editors for IEEE T-SP, T-NN and T-CSVT, T-IP and Signal Processing Magazine (SPM). He is currently on the editorial board of ZTE Communications, ETRI, IJDMB and JSPS journals. He served as the Program Co-Chair of IEEE ICME 2016 and was the Program Co-Chairs of ICASSP 1998 and ISCAS 2009. Dr. Hwang is a fellow of IEEE since 2001.


Thanks to the quick advances of deep learning in various forms of neural network architectures, such as multilayer perceptrons (MLPs), convolution neural networks (CNNs), and recurrent/graph neural networks (RNNs/GNNs), we have been enjoying the evolutionary success on various traditional machine learning tasks, such as classification, localization, detection and segmentation, perception translation/imagination, as well as experienced based structured decision making, which gradually meet/mimic the human perception (see, hear, speak, smell, feel, touch, etc) and rational/logical decision-making capabilities, resulting in so-called deep learning based AI. In this talk, I will quickly review these evolutions of approximating human intelligence from deep learning perspectives, and without loss of generality, talk about some challenges the community is trying to overcome, e.g.,

  • Open-Set Long-Tailed Recognition (OLTR)
  • Few Shot Learning and Domain Adaptation
  • Multi-Object and Multi-Camera Tracking (MTMCT)
  • Video based 3D Human Pose Estimation in the Wild
  • Recurrent/Graph Neural Networks (RNNs/GNNs)
  • Image Captioning and Scene Graph Generation
  • PointCloud Data Classification/Segmentation/Detection