Matrix Completion Under Interval Uncertainty: Highlights

Jakub Marecek; Peter Richtarik; Martin Takac

doi:10.1007/978-3-030-10997-4_38

Matrix Completion Under Interval Uncertainty: Highlights

Jakub Marecek, Peter Richtarik, Martin Takac

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We present an overview of inequality-constrained matrix completion, with a particular focus on alternating least-squares (ALS) methods. The simple and seemingly obvious addition of inequality constraints to matrix completion seems to improve the statistical performance of matrix completion in a number of applications, such as collaborative filtering under interval uncertainty, robust statistics, event detection, and background modelling in computer vision. An ALS algorithm MACO by Marecek et al. outperforms others, including Sparkler, the implementation of Li et al. Code related to this paper is available at: http://optml.github.io/ac-dc/.

Original language	English (US)
Title of host publication	Machine Learning and Knowledge Discovery in Databases
Publisher	Springer Nature
Pages	621-625
Number of pages	5
ISBN (Print)	9783030109967
DOIs	https://doi.org/10.1007/978-3-030-10997-4_38
State	Published - Jan 18 2019

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The work of JM received funding from the European Union’s Horizon 2020 Programme (Horizon2020/2014-2020) under grant agreement No. 688380. The work of MT was partially supported by the U.S. National Science Foundation, under award numbers NSF:CCF:1618717, NSF:CMMI:1663256, and NSF:CCF:1740796. PR acknowledges support from KAUST Faculty Baseline Research Funding Program.

Access to Document

10.1007/978-3-030-10997-4_38

https://repository.kaust.edu.sa/bitstream/10754/631384/1/734_Marecek.pdf

Cite this

@inproceedings{75943fc8991a4eebb4c235dd29743623,

title = "Matrix Completion Under Interval Uncertainty: Highlights",

abstract = "We present an overview of inequality-constrained matrix completion, with a particular focus on alternating least-squares (ALS) methods. The simple and seemingly obvious addition of inequality constraints to matrix completion seems to improve the statistical performance of matrix completion in a number of applications, such as collaborative filtering under interval uncertainty, robust statistics, event detection, and background modelling in computer vision. An ALS algorithm MACO by Marecek et al. outperforms others, including Sparkler, the implementation of Li et al. Code related to this paper is available at: http://optml.github.io/ac-dc/.",

author = "Jakub Marecek and Peter Richtarik and Martin Takac",

note = "KAUST Repository Item: Exported on 2020-10-01 Acknowledgements: The work of JM received funding from the European Union{\textquoteright}s Horizon 2020 Programme (Horizon2020/2014-2020) under grant agreement No. 688380. The work of MT was partially supported by the U.S. National Science Foundation, under award numbers NSF:CCF:1618717, NSF:CMMI:1663256, and NSF:CCF:1740796. PR acknowledges support from KAUST Faculty Baseline Research Funding Program.",

year = "2019",

month = jan,

day = "18",

doi = "10.1007/978-3-030-10997-4_38",

language = "English (US)",

isbn = "9783030109967",

pages = "621--625",

booktitle = "Machine Learning and Knowledge Discovery in Databases",

publisher = "Springer Nature",

}

TY - GEN

T1 - Matrix Completion Under Interval Uncertainty: Highlights

AU - Marecek, Jakub

AU - Richtarik, Peter

AU - Takac, Martin

N1 - KAUST Repository Item: Exported on 2020-10-01 Acknowledgements: The work of JM received funding from the European Union’s Horizon 2020 Programme (Horizon2020/2014-2020) under grant agreement No. 688380. The work of MT was partially supported by the U.S. National Science Foundation, under award numbers NSF:CCF:1618717, NSF:CMMI:1663256, and NSF:CCF:1740796. PR acknowledges support from KAUST Faculty Baseline Research Funding Program.

PY - 2019/1/18

Y1 - 2019/1/18

N2 - We present an overview of inequality-constrained matrix completion, with a particular focus on alternating least-squares (ALS) methods. The simple and seemingly obvious addition of inequality constraints to matrix completion seems to improve the statistical performance of matrix completion in a number of applications, such as collaborative filtering under interval uncertainty, robust statistics, event detection, and background modelling in computer vision. An ALS algorithm MACO by Marecek et al. outperforms others, including Sparkler, the implementation of Li et al. Code related to this paper is available at: http://optml.github.io/ac-dc/.

AB - We present an overview of inequality-constrained matrix completion, with a particular focus on alternating least-squares (ALS) methods. The simple and seemingly obvious addition of inequality constraints to matrix completion seems to improve the statistical performance of matrix completion in a number of applications, such as collaborative filtering under interval uncertainty, robust statistics, event detection, and background modelling in computer vision. An ALS algorithm MACO by Marecek et al. outperforms others, including Sparkler, the implementation of Li et al. Code related to this paper is available at: http://optml.github.io/ac-dc/.

UR - http://hdl.handle.net/10754/631384

UR - https://link.springer.com/chapter/10.1007%2F978-3-030-10997-4_38

UR - http://www.scopus.com/inward/record.url?scp=85061122881&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-10997-4_38

DO - 10.1007/978-3-030-10997-4_38

M3 - Conference contribution

SN - 9783030109967

SP - 621

EP - 625

BT - Machine Learning and Knowledge Discovery in Databases

PB - Springer Nature

ER -

Matrix Completion Under Interval Uncertainty: Highlights

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this