2011年
Tensor Factorization Using Auxiliary Information.
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, PT II
- ,
- ,
- ,
- 巻
- 6912
- 号
- 開始ページ
- 501
- 終了ページ
- 516
- 記述言語
- 英語
- 掲載種別
- 研究論文(国際会議プロシーディングス)
- DOI
- 10.1007/978-3-642-23783-6_32
- 出版者・発行元
- SPRINGER-VERLAG BERLIN
Most of the existing analysis methods for tensors (or multi-way arrays) only assume that tensors to be completed are of low rank. However, for example, when they are applied to tensor completion problems, their prediction accuracy tends to be significantly worse when only limited entries are observed. In this paper, we propose to use relationships among data as auxiliary information in addition to the low-rank assumption to improve the quality of tensor decomposition. We introduce two regularization approaches using graph Laplacians induced from the relationships, and design iterative algorithms for approximate solutions. Numerical experiments on tensor completion using synthetic and benchmark datasets show that the use of auxiliary information improves completion accuracy over the existing methods based only on the low-rank assumption, especially when observations are sparse.
- リンク情報
-
- DOI
- https://doi.org/10.1007/978-3-642-23783-6_32
- DBLP
- https://dblp.uni-trier.de/rec/conf/pkdd/NaritaHTK11
- Web of Science
- https://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=JSTA_CEL&SrcApp=J_Gate_JST&DestLinkType=FullRecord&KeyUT=WOS:000316551300032&DestApp=WOS_CPL
- URL
- https://dblp.uni-trier.de/conf/pkdd/2011-2
- URL
- https://dblp.uni-trier.de/db/conf/pkdd/pkdd2011-2.html#NaritaHTK11
- ID情報
-
- DOI : 10.1007/978-3-642-23783-6_32
- ISSN : 0302-9743
- DBLP ID : conf/pkdd/NaritaHTK11
- Web of Science ID : WOS:000316551300032