Reducing hubness for kernel regression

Kazuo Hara, Ikumi Suzuki, Kei Kobayashi, Kenji Fukumizu, Miloš Radovanović

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


In this paper, we point out that hubness—some samples in a high-dimensional dataset emerge as hubs that are similar to many other samples—influences the performance of kernel regression. Because the dimension of feature spaces induced by kernels is usually very high, hubness occurs, giving rise to the problem of multicollinearity, which is known as a cause of instability of regression results. We propose hubnessreduced kernels for kernel regression as an extension of a previous approach for kNN classification that reduces spatial centrality to eliminate hubness.

Original languageEnglish
Title of host publicationSimilarity Search and Applications - 8th International Conference, SISAP 2015, Proceedings
EditorsRichard Connor, Giuseppe Amato, Fabrizio Falchi, Claudio Gennaro
PublisherSpringer Verlag
Number of pages6
ISBN (Print)9783319250861
Publication statusPublished - 2015
Externally publishedYes
Event8th International Conference on Similarity Search and Applications, SISAP 2015 - Glasgow, United Kingdom
Duration: 2015 Oct 122015 Oct 14

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other8th International Conference on Similarity Search and Applications, SISAP 2015
Country/TerritoryUnited Kingdom

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Reducing hubness for kernel regression'. Together they form a unique fingerprint.

Cite this