The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
우리는 협력적 필터링 작업에 대한 일부 표준 제약 조건을 사용하여 낮은 순위 행렬 클래스의 일반화 오류 범위를 증명합니다. 우리의 경계는 순위나 관련 수량만 사용하는 알려진 경계에 비해 추가 항목을 사용하여 더 엄격합니다. L1 and L∞ 제약 조건을 고려합니다. 또한 클래스의 Rademacher 복잡성에 대한 경계가 최적임을 보여줍니다.
Ken-ichiro MORIDOMI
Kyushu University
Kohei HATANO
Kyushu University
Eiji TAKIMOTO
Kyushu University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Ken-ichiro MORIDOMI, Kohei HATANO, Eiji TAKIMOTO, "Tighter Generalization Bounds for Matrix Completion Via Factorization Into Constrained Matrices" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 8, pp. 1997-2004, August 2018, doi: 10.1587/transinf.2017EDP7339.
Abstract: 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.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDP7339/_p
부
@ARTICLE{e101-d_8_1997,
author={Ken-ichiro MORIDOMI, Kohei HATANO, Eiji TAKIMOTO, },
journal={IEICE TRANSACTIONS on Information},
title={Tighter Generalization Bounds for Matrix Completion Via Factorization Into Constrained Matrices},
year={2018},
volume={E101-D},
number={8},
pages={1997-2004},
abstract={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.},
keywords={},
doi={10.1587/transinf.2017EDP7339},
ISSN={1745-1361},
month={August},}
부
TY - JOUR
TI - Tighter Generalization Bounds for Matrix Completion Via Factorization Into Constrained Matrices
T2 - IEICE TRANSACTIONS on Information
SP - 1997
EP - 2004
AU - Ken-ichiro MORIDOMI
AU - Kohei HATANO
AU - Eiji TAKIMOTO
PY - 2018
DO - 10.1587/transinf.2017EDP7339
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2018
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.
ER -