Semi-supervised regression using cluster ensemble and low-rank co-association matrix decomposition under uncertainties

Vladimir Berikov, Alexander Litvinenko

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

In this paper, we solve a semi-supervised regression problem. Due to the luck of knowledge about the data structure and the presence of random noise, the considered data model is uncertain. We propose a method which combines graph Laplacian regularization and cluster ensemble methodologies. The co-association matrix of the ensemble is calculated on both labeled and unlabeled data; this matrix is used as a similarity matrix in the regularization framework to derive the predicted outputs. We use the low-rank decomposition of the co-association matrix to significantly speedup calculations and reduce memory. Numerical experiments using the Monte Carlo approach demonstrate robustness, efficiency, and scalability of the proposed method.

Original languageEnglish
Title of host publicationProceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019
EditorsM. Papadrakakis, V. Papadopoulos, G. Stefanou
PublisherNational Technical University of Athens
Pages229-242
Number of pages14
ISBN (Print)9786188284494
DOIs
Publication statusPublished - 2019
Event3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019 - Crete, Greece
Duration: 24 Jun 201926 Jun 2019

Publication series

NameProceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019

Conference

Conference3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019
CountryGreece
CityCrete
Period24.06.201926.06.2019

Keywords

  • Cluster ensemble
  • Co-association matrix
  • Graph Laplacian regularization
  • Hierarchical matrices
  • Low-rank matrix decomposition
  • Semi-supervised regression

Fingerprint Dive into the research topics of 'Semi-supervised regression using cluster ensemble and low-rank co-association matrix decomposition under uncertainties'. Together they form a unique fingerprint.

Cite this