Stochastic cluster embedding

DOI10.1007/S11222-022-10186-ZzbMATH Open1499.62040arXiv2108.08003OpenAlexW4311375103MaRDI QIDQ2104017FDOQ2104017

Jukka Corander, Samuel Kaski, Denis Sedov, Zhirong Yang, Yu-Wei Chen

Publication date: 9 December 2022

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: Neighbor Embedding (NE) aims to preserve pairwise similarities between data items and has been shown to yield an effective principle for data visualization. However, even the best existing NE methods such as Stochastic Neighbor Embedding (SNE) may leave large-scale patterns hidden, for example clusters, despite strong signals being present in the data. To address this, we propose a new cluster visualization method based on the Neighbor Embedding principle. We first present a family of Neighbor Embedding methods that generalizes SNE by using non-normalized Kullback-Leibler divergence with a scale parameter. In this family, much better cluster visualizations often appear with a parameter value different from the one corresponding to SNE. We also develop an efficient software that employs asynchronous stochastic block coordinate descent to optimize the new family of objective functions. Our experimental results demonstrate that the method consistently and substantially improves the visualization of data clusters compared with the state-of-the-art NE approaches.

Full work available at URL: https://arxiv.org/abs/2108.08003

Recommendations

• Visualizing data using t-SNE
• Clustering with t-SNE, provably
• Parametric Embedding for Class Visualization
• scientific article
• Using global t-SNE to preserve intercluster data structure

zbMATH Keywords

clusteringvisualizationnonlinear dimensionality reductionstochastic optimizationinformation divergenceneighbor embedding

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08)

Cites Work

• Title not available (Why is that?)
• Differential-geometrical methods in statistics
• Information retrieval perspective to nonlinear dimensionality reduction for data visualization
• Iteration complexity of randomized block-coordinate descent methods for minimizing a composite function
• Title not available (Why is that?)

Cited In (1)

• sce

Uses Software

• largeVis
• t-SNE
• ELFI
• UMAP
• LargeVis
• DeepView

This page was built for publication: Stochastic cluster embedding

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2104017)