How to implement the Bayes’ formula in the age of ML?
组织者
演讲者
孙泽钜
时间
2024年12月04日 14:00 至 15:00
地点
Online
摘要
This talk is based on a recent report of A. Taghvaei and P. G. Mehta on Bayes filtering. It includes a historical survey of the traditional filtering algorithms, a discussion of the fundamental limitations of these algorithms in non-Gaussian and high-dimensional applications, a new optimization formulation of the Bayes’ law based on optimal transportation theory. The relationship of this new formulation and the feedback particle filter in continuous-time settings of the nonlinear filtering problem will also be illustrated.