ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 A Distributed Support Vector Machine Learning Over Wireless Sensor Networks
Cited 40 time in scopus Download 25 time Share share facebook twitter linkedin kakaostory
저자
김우진, Milos S. Stankovi´, Karl H. Johansson, 김현진
발행일
201511
출처
IEEE Transactions on Cybernetics, v.45 no.11, pp.2599-2611
ISSN
2168-2267
출판사
IEEE
DOI
https://dx.doi.org/10.1109/TCYB.2014.2377123
협약과제
14MC1100, SMART Post 구축 기술 개발, 정훈
초록
This paper is about fully-distributed support vector machine (SVM) learning over wireless sensor networks. With the concept of the geometric SVM, we propose to gossip the set of extreme points of the convex hull of local data set with neighboring nodes. It has the advantages of a simple communication mechanism and finite-time convergence to a common global solution. Furthermore, we analyze the scalability with respect to the amount of exchanged information and convergence time, with a specific emphasis on the small-world phenomenon. First, with the proposed naive convex hull algorithm, the message length remains bounded as the number of nodes increases. Second, by utilizing a small-world network, we have an opportunity to drastically improve the convergence performance with only a small increase in power consumption. These properties offer a great advantage when dealing with a large-scale network. Simulation and experimental results support the feasibility and effectiveness of the proposed gossip-based process and the analysis.
KSP 제안 키워드
Communication Mechanism, Convergence performance, Convex hull algorithm, Data sets, Exchanged information, Extreme points, Finite-time convergence, Global solution, Gossip-based, Large-scale network, Power Consumption