File(s) under permanent embargo
Matrix factorization for identifying noisy labels of multi-label instances
conference contribution
posted on 2018-01-01, 00:00 authored by X Chen, G Yu, C Domeniconi, J Wang, Zili ZhangZili Zhang© Springer International Publishing AG, part of Springer Nature 2018. Current effort on multi-label learning generally assumes that the given labels are noise-free. However, obtaining noise-free labels is quite difficult and often impractical. In this paper, we study how to identify a subset of relevant labels from a set of candidate ones given as annotations to instances, and introduce a matrix factorization based method called MF-INL. It first decomposes the original instance-label association matrix into two low-rank matrices using nonnegative matrix factorization with feature-based and label-based constraints to retain the geometric structure of instances and label correlations. MF-INL then reconstructs the association matrix using the product of the decomposed matrices, and identifies associations with the lowest confidence as noisy associations. An empirical study on real-world multi-label datasets with injected noisy labels shows that MF-INL can identify noisy labels more accurately than other related solutions and is robust to input parameters. We empirically demonstrate that both feature-based and label-based constraints contribute to boosting the performance of MF-INL.
History
Event
Artificial Intelligence. Pacific Rim International Conference (16th : 2018 : Nanjing, China)Volume
11013Series
Lecture Notes in Computer SciencePagination
508 - 517Publisher
SpringerLocation
Nanjing, ChinaPlace of publication
Cham, SwitzerlandPublisher DOI
Start date
2018-08-28End date
2018-08-31ISSN
0302-9743eISSN
1611-3349ISBN-13
9783319973098Publication classification
E1.1 Full written paper - refereedCopyright notice
2018, Springer International Publishing AGEditor/Contributor(s)
X Geng, B KangTitle of proceedings
PRICAI 2018: Proceedings of the 16th Pacific Rim International Conference on Artificial Intelligence : Trends in Artificial IntelligenceUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC