Prof Minho Lee

Beyond Perception, Toward Cognition

- Brain inspired deep learning -

By Prof. Minho Lee

School of Electronics Engineering, Kyungpook National University, Korea

Abstract:

For recent years, the most prominent subfield of machine learning has been deep learning. Deep learning is to train many hidden layers in deep feedforward and feedback neural networks, which is inspired by the basic principles of brain information processing mechanism. It has taken a place not only in many professional academic areas such as computer vision, speech recognition and language processing, but also in industrial arena. Recent breakthroughs have been introducing and adapting, deep learning methods eventually defeating many classical state-of-art algorithms in many engineering application fields related with “Perception”. However, it still has many limits in real world application including data collection and labeling issues as well as generalization, stability and plasticity dilemma. In this talk, I will explain some of efforts to solve the current limitation of deep learning for perception tasks, and a possibility to develop a next generation of current deep learning methods for implementing human-level cognition system, so called Cognitron, such as intention and emotion understanding as well as advanced natural language understanding. Also, I will show some of practical applications for implementing digital companion in AI flagship project.

Bio:

Prof. Minho Lee received his Ph.D. from the Korea Advanced Institute of Science and Technology (KAIST) in 1995, and is currently a Professor in the School of Electronics Engineering, Kyungpook National University and directors for AI institute of technology and KNU-LG Convergence Research Center, Daegu, Korea. He established Mobile Technology Commercial Center at Daegu, and worked for Education & Training Department as a director from 2005 to 2006. Also, he was a visiting professor for Dept. of Brain and Cognitive Science at MIT from 2006 to 2007. He received several best paper awards at international conferences including ICONIP (2007 and 2009), IDEAL (2008), ICAISC (2006), and ISABEL (2006), best award for industry collaboration at KNU (2014) and Excellent Service Award from APNNA (2014). He has worked with several international journals (including Neural Networks and Nature Intelligence) as an Associate Editor, and also for international conferences as program chairs and special session chairs. Dr. Lee was President of the Asia-Pacific Neural Network Assembly (APNNA) and General Chair for the International Conference on Neural Information Processing (ICONIP) in 2013. Also, he is a General Chair for International Conference on Human Agent Interaction (HAI) in 2015. His research interests include brain-neuroinformatics, biologically-inspired vision systems, human augmented cognition, selective attention, brain-machine interaction, and intelligent sensor systems.