Speaker : Song Chong, ICT Endowed Chair Professor, KAIST AI&EE, Korea
Venue : Online by Zoom
Date & time : Wednessday 8th Dec 2021, 16:00~17:00
Abstract : Reinforcement learning, an Artificial Intelligence method for continuous decisions, is described. It helps you understand reinforcement learning by analyzing how basic reinforcement learning works. As an example of reinforcement learning, the Deep-Q network is described, and how deep reinforcement learning is developing is presented through a graph. We share real-world examples of advanced reinforcement learning applied in our lives. Through sharing these real-world cases, we analyze how reinforcement learning should be developed in the future. Finally, a seminar will be held to present the current status of Korea's AI competitiveness and the future direction of Korea's artificial intelligence.
Keywords : Reinforcement learning, Deep-Q-Networks, Inverse Reinforcement learning
Bio :
Degree
Ph.D. (1995) Univ.of Texas at Austin
Achievement
- IEEE William R. Bennett Prize in 2013
- IEEE SECON Best Paper Award in 2013
- “On the Levy-Walk Nature of Human Mobility,” IEEE/ACM Trans. on Networking, 2011
- “SLAW: A New Mobility Model for Human Walks,” IEEE/ACM Trans. on Networking, 2012
- "Mobile Data Offloading: How Much Can WiFi Deliver?," IEEE/ACM Trans. on Networking, 2013