- Friday-Saturday October 22-23, 2017
- Nogodan Guest House and Hotel, Jeonnam, Korea
With the incredible success of deep learning, machine learning and artificial intelligence have attracted great interest from the public. The success is widely attributed to several mathematically ingenious algorithms and theories. However, it is also true that to some extent the success relies on the advance of the computing hardwares. Some features remain mysteries, and fully understanding them may bring further advancement of the machine learning. On top of it, its has recently been shown that the deep network architecture shares a common mathematical structure with the tensor network states, which have been developed independently in the field of quantum information theory and condensed matter physics.
The Quantum Control Lab, lead by Prof. Mahn-Soo Choi, is studying the mathematical structure of the deep networks and the tensor network states to reveal the unknown deep relation between the two fields. They are aiming to the develop a machine learning architecture which is specialized for the solution of strongly correlated systems, which are known to be very hard to simulate in conventional methods with classical computers.
Program
- Friday Sep 21, 2017 Registration
- 18:00- Registration
- Friday Sep 22, 2017 Deep Learning
- 09:00-09:50 Mahn-Soo Choi
- 10:00-10:50 Chang-Woo Lee
- 11:00-11:50 Ho-Jun Kim
- 12:00-14:00 Lunch
- Friday Sep 22, 2017 Tensor Network States
- 14:00-14:50 Ho-Joon Kim I
- 15:00-15:50 Ho-Joon Kim II
- 16:00-16:50 MInchul Lee
- 17:00-17:50 Mahn-Soo Choi
- 18:00- 20:00 Dinner
- Friday Sep 23, 2017 Research Planning Forum
- 09:00-09:50 Seungju Han
- 10:00-10:50 Seung Park
- 11:00-11:50 Ha-Eum Kim
- 11:50-12:00 Closing
- 12:00-13:30 Lunch
Deep Learning
Chairman: Ho-Joon Kim
- Choi, Mahn-Soo (Korea Univ), Introduction to Machine Learning
- Lee, Chang-Woo (KIAS), Introduction to Deep Learning
- Kim, Ho-Joon (Pusan Nat'l Univ), Representation Theorem in Artificial Neural Network
Tensor Network States
Chairman: Mahn-Soo Choi
- Kim, Ho-Joon (Pusan Nat'l Univ), Introduction to Tensor Network States
- Kim, Ho-Joon (Pusan Nat'l Univ), Entanglement Renormalization with Tensor Network States
- Lee, Minchul (Kyunghee Univ), Introduction to DMRG and NRG Methods
- Choi, Mahn-Soo (Korea Univ), Common Mathematical Structures for Tensor Network States and Deep Learning
Research Panning Forum
Chairman: Mahn-Soo Choi
- Panel Members: Mahn-Soo Choi (Korea Univ), Ho-Jun Kim (Pusan Nat'l Univ), Chang-Woo Lee (KIAS), Minchul Lee (Kyunghee Univ)
Organizer
- The Quantum Control Lab. (Korea University)
Sponsors
- Korea University Future Research Grant (KU FRG).
- BK21 Plus Initiative of Department of Physics at Korea University.