TY - JOUR
T1 - Tighter generalization bounds for matrix completion via factorization into constrained matrices
AU - Moridomi, Ken ichiro
AU - Hatano, Kohei
AU - Takimoto, Eiji
N1 - Funding Information:
This work is supported in part by JSPS KAKENHI Grant
Funding Information:
This work is supported in part by JSPS KAKENHI Grant Number 16K00305 and JSPS KAKENHI Grant Number 15H02667, respectively.
PY - 2018/8
Y1 - 2018/8
N2 - We prove generalization error bounds of classes of low-rank matrices with some norm constraints for collaborative filtering tasks. Our bounds are tighter, compared to known bounds using rank or the related quantity only, by taking the additional L1 and L∞ constraints into account. Also, we show that our bounds on the Rademacher complexity of the classes are optimal.
AB - We prove generalization error bounds of classes of low-rank matrices with some norm constraints for collaborative filtering tasks. Our bounds are tighter, compared to known bounds using rank or the related quantity only, by taking the additional L1 and L∞ constraints into account. Also, we show that our bounds on the Rademacher complexity of the classes are optimal.
UR - http://www.scopus.com/inward/record.url?scp=85052026888&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85052026888&partnerID=8YFLogxK
U2 - 10.1587/transinf.2017EDP7339
DO - 10.1587/transinf.2017EDP7339
M3 - Article
AN - SCOPUS:85052026888
SN - 0916-8532
VL - E101D
SP - 1997
EP - 2004
JO - IEICE Transactions on Information and Systems
JF - IEICE Transactions on Information and Systems
IS - 8
ER -