Location via proxy:
[ UP ]
[Report a bug]
[Manage cookies]
No cookies
No scripts
No ads
No referrer
Show this form
CORE
🇺🇦
make metadata, not war
Services
Services overview
Explore all CORE services
Access to raw data
API
Dataset
FastSync
Content discovery
Recommender
Discovery
OAI identifiers
OAI Resolver
Managing content
Dashboard
Bespoke contracts
Consultancy services
Support us
Support us
Membership
Sponsorship
Community governance
Advisory Board
Board of supporters
Research network
About
About us
Our mission
Team
Blog
FAQs
Contact us
Riemannian pursuit for big matrix recovery
Authors
SJ Pan
M Tan
+3 more
IW Tsang
B Vandereycken
L Wang
Publication date
1 January 2014
Publisher
Abstract
Copyright © (2014) by the International Machine Learning Society (IMLS) All rights reserved. Low rank matrix recovery is a fundamental task in many real-world applications. The perfor-mance of existing methods, however, deteriorates significantly when applied to ill-conditioned or large-scale matrices. In this paper, we therefore propose an efficient method, called Riemannian Pursuit (RP), that aims to address these two problems simultaneously. Our method consists of a sequence of fixed-rank optimization problems. Each subproblem, solved by a nonlinear Rieman-nian conjugate gradient method, aims to correct the solution in the most important subspace of increasing size. Theoretically, RP converges linearly under mild conditions and experimental results show that it substantially outperforms existing methods when applied to large-scale and ill-conditioned matrices
Similar works
Full text
Available Versions
OPUS - University of Technology Sydney
See this paper in CORE
Go to the repository landing page
Download from data provider
oai:opus.lib.uts.edu.au:10453/...
Last time updated on 18/10/2019