Seminar by Prof. G. George YIN from Department of Mathematics, University of Connecticut

Nonlinear filtering is a fundamental problem in information theory, communication, signal processing, control and optimization, and systems theory. In the 1960s, celebrated results on nonlinear filtering were obtained. Nevertheless, the computational issues for nonlinear filtering remained to be a long-standing and challenging problem. In this talk, in lieu of treating an infinite dimensional problem for obtaining the conditional distribution, or conditional measure, we construct finite-dimensional approximations using deep neural networks for the optimal weights. Two recursions are used in the algorithm. One of them is the approximation of the optimal weight and the other is for approximating the optimal learning rate.

This presentation is based on a joint work with Qing Zhang (University of Georgia), and Hongjiang Qian (University of Connecticut).

George Yin is a Professor at Department of Mathematics, University of Connecticut. He obtained PhD degree in Applied Mathematics in 1987 from Brown University. He then joined the Wayne State University and became a full professor in 1996. His research interests include Stochastic Systems Theory and Numerical Method, Stochastic Optimization, Control, and Approximation, Mathematics of Biology and System Identification.

Prof. Yin was elected as Fellow of IEEE in 2002 for contributions to approximation, optimization, and control of stochastic systems. He is also Fellow of IFAC and a Fellow of SIAM.