AM Seminar: More Optimal Way to Optimize via Fractional Calculus Beyond Nesterov

Speaker Name: 
YangQuan Chen
Speaker Title: 
Professor
Speaker Organization: 
UC Merced
Start Time: 
Monday, February 11, 2019 - 4:00pm
End Time: 
Monday, February 11, 2019 - 5:00pm
Location: 
BE 372
Organizer: 
Abhishek Halder

Abstract

It is now more and more widely accepted that for iterative optimization, it is logical to ask "What is the optimal way to optimize?" Nesterov accelerated gradient method (AGM) is introduced first and then Michael I. Jordan’s ICM18 talk is briefly reviewed. We re-examine the iteration-axis dynamics from control system point of view using z-transform and show that Nesterov scheme is a second order dynamic system along iteration axis. In continuum limit, continuous-time system theory can be used for designing new iteration schemes. As an example, we introduce a continuous-time fractional order filter and then discretize it to form a new iterative scheme, resulting a potentially more optimal way to optimize. Stochastic AGM will also be discussed briefly with an indication of a potential benefit of using fractional order stochasticity in achieving more optimal way to optimize. It is hoped that this talk will open new research opportunities in machine learning in general, and deep artificial neural network learning in particular.

Bio

YangQuan Chen earned his Ph.D. from Nanyang Technological University, Singapore, in 1998. He had been a faculty of Electrical Engineering at Utah State University from 2000-12. He joined the School of Engineering, University of California, Merced in summer 2012 teaching "Mechatronics", "Engineering Service Learning" and "Unmanned Aerial Systems" for undergraduates; "Fractional Order Mechanics" and "Nonlinear Controls" for graduates. His research interests include mechatronics for sustainability, cognitive process control, small multi-UAV based cooperative multi-spectral "personal remote sensing", applied fractional calculus in controls, modeling and complex signal processing; distributed measurement and control of distributed parameter systems with mobile actuator and sensor networks.

Dr. Chen serves as a Co-Chair for IEEE Robotics and Automation Society Technical Committee (TC) on Unmanned Aerial Vehicle and Aerial Robotics (12-18). He recently served the TC Chair for the ASME DED Mechatronics Embedded Systems Applications (2009-10); Associated Editor (AE) for IEEE Trans. on Control Systems Technology (00-16), ISA Trans. (12-17), IFAC Control Engineering Practice (12-17), IET Control Theory and Applications (15-18) and Journal of Dynamics Systems, Measurements and Control (09-15). He now serves as Topic Editor-in-Chief of International Journal of Advanced Robotic Systems (Field Robotics), Section AE (Remote Sensors) for Sensors, Senior Editor for International Journal of Intelligent Robotic Systems, Topical AE for Nonlinear Dynamics (18-) and AE for IFAC Mechatronics, Intelligent Service Robotics, Energy Sources (Part A) (18-) and Fractional Calculus and Applied Analysis. He is a member of IEEE, ASME, AIAA, ASPRS, AUVSI and AMA. He relies on Google citation page to keep track of his publications at https://scholar.google.com/citations?user=RDEIRbcAAAAJ

Dr. Chen started some new investigations, published some papers, patents and books, graduated some students, hosted some visiting scholars and also received some awards including the IFAC World Congress Best Journal Paper Award (Control Engineering Practice, 2011), First Place Awards for 2009 and 2011 AUVSI SUAS competitions, and most importantly, the "Relationship Counselor" award from IEEE Utah State University Student Branch for "explaining human relationship using control theory". His is listed in Highly Cited Researchers by Clarivate Analytics in 2018.