ORBi: Liao Yongjun - Network Distance Prediction Based on Decentralized Matrix Factorization

Reference : Network Distance Prediction Based on Decentralized Matrix Factorization


	Document type :	Scientific congresses and symposiums : Paper published in a journal
	Discipline(s) :	Engineering, computing & technology : Computer science
	To cite this reference:	http://hdl.handle.net/2268/36473

Title :	Network Distance Prediction Based on Decentralized Matrix Factorization
Language :	English
Author, co-author :	Liao, Yongjun [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Réseaux informatiques >]
	Geurts, Pierre [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
	Leduc, Guy [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Réseaux informatiques >]
Publication date :	11-May-2010
Journal title :	Lecture Notes in Computer Science
Publisher :	Springer
Volume :	6091
Pages :	15-26
On invitation :	No
Audience :	International
ISSN :	0302-9743
e-ISSN :	1611-3349
City :	Berlin
Country :	Germany
Event name :	IFIP Networking 2010
Event date :	11-13 May 2010
Event place (city) :	Chennai
Event country :	India
Keywords :	[en] Network coordinate system ; matrix factorization
Abstract :	[en] Network Coordinate Systems (NCS) are promising techniques to predict unknown network distances from a limited number of measurements. Most NCS algorithms are based on metric space embedding and suffer from the inability to represent distance asymmetries and Triangle Inequality Violations (TIVs). To overcome these drawbacks, we formulate the problem of network distance prediction as guessing the missing elements of a distance matrix and solve it by matrix factorization. A distinct feature of our approach, called Decentralized Matrix Factorization (DMF), is that it is fully decentralized. The factorization of the incomplete distance matrix is collaboratively and iteratively done at all nodes with each node retrieving only a small number of distance measurements. There are no special nodes such as landmarks nor a central node where the distance measurements are collected and stored. We compare DMF with two popular NCS algorithms: Vivaldi and IDES. The former is based on metric space embedding, while the latter is also based on matrix factorization but uses landmarks. Experimental results show thatDMF achieves competitive accuracy with the double advantage of having no landmarks and of being able to represent distance asymmetries and TIVs.
Funders :	EU, and Belgian State Science Policy Office
Name of the research project :	FP7 ECODE, NoE PASCAL2, IAP DYSCO
Target :	Researchers
Permalink :	http://hdl.handle.net/2268/36473
DOI :	10.1007/978-3-642-12963-6_2
Commentary :	Best paper award
Framework Programme / European Project :	FP7 ; 223936 - ECODE - Experimental COgnitive Distributed Engine

File(s) associated to this reference

Fulltext file(s):

	File	Commentary	Version	Size	Access
Open access	YL-networking2010.pdf		Author postprint	1.71 MB	View/Open

All documents in ORBi are protected by a user license.

Dspace v1.5