Parametric t-distributed stochastic exemplar-centered embeddin

Download
  1. (PDF, 6 MB)
  2. Get@NRC: Parametric t-distributed stochastic exemplar-centered embeddin (Opens in a new window)
AuthorSearch for: ; Search for: ; Search for:
TypeArticle
Journal titleComputer Science
Article numberarXiv:1710.05128
Pages# of pages: 10
Physical descriptionversion 2
AbstractParametric embedding methods such as parametric t-SNE (pt-SNE) have been widely adopted for data visualization and out-of-sample data embedding without further computationally expensive optimization or approximation. However, the performance of pt-SNE is highly sensitive to the hyper-parameter batch size due to conflicting optimization goals, and often produces dramatically different embeddings with different choices of user-defined perplexities. To effectively solve these issues, we present parametric t-distributed stochastic exemplar-centered embedding methods. Our strategy learns embedding parameters by comparing given data only with precomputed exemplars, resulting in a cost function with linear computational and memory complexity, which is further reduced by noise contrastive samples. Moreover, we propose a shallow embedding network with high-order feature interactions for data visualization, which is much easier to tune but produces comparable performance in contrast to a deep neural network employed by pt-SNE. We empirically demonstrate, using several benchmark datasets, that our proposed methods significantly outperform pt-SNE in terms of robustness, visual effects, and quantitative evaluations.
Publication date
PublisherCornell University Library
Linkhttps://arxiv.org/abs/1710.05128
LanguageEnglish
AffiliationDigital Technologies; National Research Council Canada
Peer reviewedNo
NPARC number23002468
Export citationExport as RIS
Report a correctionReport a correction
Record identifiercd03cba4-a957-41fa-9c40-15d5466274ea
Record created2017-11-15
Record modified2017-11-15
Bookmark and share
  • Share this page with Facebook (Opens in a new window)
  • Share this page with Twitter (Opens in a new window)
  • Share this page with Google+ (Opens in a new window)
  • Share this page with Delicious (Opens in a new window)
Date modified: