Decentralized Discrete-Time Algorithm for Estimating AlgebraicConnectivity of Multiagent Networks
- 수학전공
- 조회수1605
- 2016-10-25
▣ 연 사 : Kento Endo and Norikazu Takahashi
( Okayama University )
▣ 일 시 : 2016년 10월 28일(금) 3:00~4:00
▣ 장 소 : 32356A호(수학과 세미나실)
▣ 대 상 : 수학과 학부생 및 대학원생
▣ 다 과 : 2시 45분부터
Abstract
In this talk, we present a new decentralized discrete-time algorithm for each agent in a network to estimate the algebraic connectivity of the network. The algebraic connectivity, which is defined as the second smallest eigenvalue of the Laplacian matrix of the network, represents how strongly the network is connected. The word "decentralized" means that no central agent exists in the network and thus each agent has to work on the task by communicating only with its neighboring agents. Hence the challenge of this study is to design an algorithm (communication protocol) so that each agent can properly find the global information of the network through local communication. In this talk, we begin with a brief introduction to multiagent networks and algebraic connectivity. We then explain the proposed algorithm in detail, and present theoretical and experimental results to show its validity. We finally show some problems to solve in order to fully understand the dynamical behavior of the proposed algorithm.
콜로퀴움 인정!!
- 이전글
- 이전글이 없습니다.