Recent Developments in Deep Learning Filtering Algorithms
组织者
演讲者
孙泽钜
时间
2023年11月10日 21:00 至 21:30
地点
Online
摘要
In this talk, we will review two kinds of deep learning filtering algorithms proposed recently: the deep Kalman filer and deep learning based optimal transportation filter. For the deep Kalman filter, the authors address the issue in the deep learning mathematical foundation and give a convergence analysis for Gaussian signal processes. For the optimal transportation filter, the authors discuss its property in high-dimensional setting and show that the optimal transportation filter can be conducted by a neural network.