Modular Dimensionality Reduction

Henry W J Reeve*, Tingting Mu, Gavin Brown

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

12 Downloads (Pure)

Abstract

We introduce an approach to modular dimensionality reduction, allowing efficient learning of multiple complementary representations of the same object. Modules are trained by optimising an unsupervised cost function which balances two competing goals: Maintaining the inner product structure within the original space, and encouraging structural diversity between complementary representations. We derive an efficient learning algorithm which outperforms gradient based approaches without the need to choose a learning rate. We also demonstrate anintriguing connection with Dropout. Empirical results demonstrate the efficacy of the method for image retrieval and classification.
Original languageEnglish
Title of host publicationProceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2018
PublisherSpringer
Pages605-619
Number of pages15
ISBN (Electronic)978-3-030-10925-7
ISBN (Print)978-3-030-10924-0
DOIs
Publication statusPublished - 18 Jan 2019
EventEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - Dublin, Ireland
Duration: 10 Sep 201814 Sep 2018
http://www.ecmlpkdd2018.org/

Publication series

NameLecture Notes in Computer Science
PublisherSpringer, Cham
Volume11051
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Abbreviated titleECML PKDD 2018
CountryIreland
CityDublin
Period10/09/1814/09/18
Internet address

Keywords

  • Ensemble learning
  • Dimensionality reduction
  • Dropout
  • Kernel principal components analysis

Fingerprint Dive into the research topics of 'Modular Dimensionality Reduction'. Together they form a unique fingerprint.

  • Cite this

    Reeve, H. W. J., Mu, T., & Brown, G. (2019). Modular Dimensionality Reduction. In Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2018 (pp. 605-619). (Lecture Notes in Computer Science; Vol. 11051). Springer. https://doi.org/10.1007/978-3-030-10925-7_37