BIMSA >
Seminar on Control Theory and Nonlinear Filtering
Deep Filtering and Stochastic Approximation Approach
Deep Filtering and Stochastic Approximation Approach
Organizer
Speaker
Zeju Sun
Time
Wednesday, June 12, 2024 10:30 AM - 11:00 AM
Venue
理科楼A-203
Abstract
In this talk, we will review some recent advances in nonlinear filtering by H. Qian, G. Yin and Q. Zhang. The first filtering algorithm is based on a deep learning framework, which can be used to deal with challenging filtering problems involving diffusions with randomly-varying switching. The second filtering algorithm is based on a recursive filtering equation in which the gain matrix is a matrix-valued parameter to be determined. This approach can deal with general filtering systems with degenerate observation noises.