Perplexity-free Parametric t-SNE

Crecchi, Francesco, de Bodt, Cyril, Verleysen, Michel, Lee, John A., Bacciu, Davide

arXiv.org Machine Learning 

The t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm is a ubiquitously employed dimensionality reduction (DR) method. Its non-parametric nature and impressive efficacy motivated its parametric extension. It is however bounded to a user-defined perplexity parameter, restricting its DR quality compared to recently developed multi-scale perplexity-free approaches. This paper hence proposes a multi-scale parametric t-SNE scheme, relieved from the perplexity tuning and with a deep neural network implementing the mapping. It produces reliable embeddings with out-of-sample extensions, competitive with the best perplexity adjustments in terms of neighborhood preservation on multiple data sets.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found