Mutual-Information-Graph Regularized Sparse Transform for Unsupervised Feature Learning

Songlin Du, Takeshi Ikenaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Unsupervised feature learning is attracting more and more attention in machine learning and computer vision because of the increasing demand for effective representation of large-scale unlabeled data in real-world applications. This paper proposes a mutual-information-graph regularized sparse transform (MIST) algorithm by taking both of feature sparsity and underlying manifold structure of observation data into consideration. The feature transform is formulated by a transform kernel and a bias matrix. To obtain feature sparsity, the sparse filtering is utilized as nonlinear activation function. A mutual information graph is proposed to describe the underlying manifold structure of the observation data. The transform kernel and the bias matrix are finally learned under the regularization of the mutual information graph. The proposed approach has both the properties of sparsity and local-structure-preservation. These two properties guarantee the discriminative power and robustness in practical applications. Experimental results on handwritten digits recognition show that the proposed approach achieves high performance compared with existing unsupervised feature learning models.

Original languageEnglish
Title of host publicationISPACS 2018 - 2018 International Symposium on Intelligent Signal Processing and Communication Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages215-219
Number of pages5
ISBN (Electronic)9781538657713
DOIs
Publication statusPublished - 2018 Nov
Event2018 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2018 - Ishigaki Island, Okinawa, Japan
Duration: 2018 Nov 272018 Nov 30

Publication series

NameISPACS 2018 - 2018 International Symposium on Intelligent Signal Processing and Communication Systems

Conference

Conference2018 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2018
CountryJapan
CityIshigaki Island, Okinawa
Period18/11/2718/11/30

Keywords

  • graph regularization
  • mutual information graph
  • sparse transform
  • Unsupervised feature learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Signal Processing

Fingerprint Dive into the research topics of 'Mutual-Information-Graph Regularized Sparse Transform for Unsupervised Feature Learning'. Together they form a unique fingerprint.

  • Cite this

    Du, S., & Ikenaga, T. (2018). Mutual-Information-Graph Regularized Sparse Transform for Unsupervised Feature Learning. In ISPACS 2018 - 2018 International Symposium on Intelligent Signal Processing and Communication Systems (pp. 215-219). [8923407] (ISPACS 2018 - 2018 International Symposium on Intelligent Signal Processing and Communication Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISPACS.2018.8923407