An invited talk by Prof. Yiru Zhang, Assistant Professor, Cergy Pontoise University, France, will be held on December 27, 2022, at Zoom (online). Prof. Yiru Zhang will share some interesting ideas about the theory of belief functions.

Date: 27 Dec 2022

Time: 04:15 PM to 05:45 PM (JST)

CO-ORGANIZED BY:

IEEE Muroran Institute of Technology Student Branch

IEEE Sapporo Section Young Professionals (YP)

Emerging Networks and Systems Laboratory (ENeS), Muroran Institute of Technology

Location:

https://zoom.us/j/7701243471?pwd=ZFcrcmNPbmxtRC9qbTlackc3bWdtZz09

Abstract:

In science and engineering it is often necessary to reason with partial knowledge and uncertain information (from sensors, experts, models, etc.) Based on the nature of the cause, the uncertainty is generally categorized as aleotoric and epistemic. Bayesian probability theory is the mostly valid mathematical tool for aleotoric uncertain reasoning but may bring unexpected issues facing epistemic one. The theory of belief functions, also known as Dempster-Shafer theory of evidence theory, is able to reason the two types of uncertainty in a unified framework by extending probability theory with set theory. In this talk, starting with a brief review of the different types of uncertainty, we will introduce how the theory of belief functions deal with them, followed by some representative applications in information fusion and machine learning.