GPGPU Linear Complexity t-SNE Optimization

Nicola Pezzotti, Julian Thijssen, Alexander Mordvinstev, Thomas Hollt, Baldur Van Lew, Boudewijn Lelieveldt, Elmar Eisemann, Anna Vilanova

Research output: Contribution to journalArticleScientificpeer-review

19 Citations (Scopus)
140 Downloads (Pure)


In recent years the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm has become one of the most used and insightful techniques for exploratory data analysis of high-dimensional data. It reveals clusters of high-dimensional data points at different scales while only requiring minimal tuning of its parameters. However, the computational complexity of the algorithm limits its application to relatively small datasets. To address this problem, several evolutions of t-SNE have been developed in recent years, mainly focusing on the scalability of the similarity computations between data points. However, these contributions are insufficient to achieve interactive rates when visualizing the evolution of the t-SNE embedding for large datasets. In this work, we present a novel approach to the minimization of the t-SNE objective function that heavily relies on graphics hardware and has linear computational complexity. Our technique decreases the computational cost of running t-SNE on datasets by orders of magnitude and retains or improves on the accuracy of past approximated techniques. We propose to approximate the repulsive forces between data points by splatting kernel textures for each data point. This approximation allows us to reformulate the t-SNE minimization problem as a series of tensor operations that can be efficiently executed on the graphics card. An efficient implementation of our technique is integrated and available for use in the widely used Google TensorFlow.js, and an open-source C++ library.
Original languageEnglish
Article number8811606
Pages (from-to)1172-1181
Number of pages10
JournalIEEE Transactions on Visualization and Computer Graphics
Issue number1
Publication statusPublished - 2020

Bibliographical note

Accepted author manuscript


  • High Dimensional Data
  • Dimensionality Reduction
  • Progressive Visual Analytics
  • Approximate Computation


Dive into the research topics of 'GPGPU Linear Complexity t-SNE Optimization'. Together they form a unique fingerprint.

Cite this