IEEE VIS 2024 Content: Hypertrix: An indicatrix for high-dimensional visualizations

Best Paper Award

Hypertrix: An indicatrix for high-dimensional visualizations

Shivam Raval - Harvard University, Boston, United States

Fernanda Viegas - Harvard University, Cambridge, United States. Google Research, Cambridge, United States

Martin Wattenberg - Harvard University, Cambridge, United States. Google Research, Cambridge, United States

Screen-reader Accessible PDF

Room: Bayshore I + II + III

2024-10-15T15:10:00Z GMT-0600 Change your timezone on the schedule page
2024-10-15T15:10:00Z
Exemplar figure, described by caption below
Hypertrix is an indicatrix for visualizing distortions in high-dimensional data projections. It is an overlay of colored elliptical glyphs on data projections, revealing both the magnitude and direction of local distortions. The hypertrix for a t-SNE projection of the MNIST dataset reveals the compactness of the digit '1' cluster with respect to other clusters.
Fast forward
Keywords

Dimensionality Reduction, High-dimensional data—Distortion—Text Visualization, Clustering

Abstract

Visualizing high dimensional data is challenging, since any dimensionality reduction technique will distort distances. A classic method in cartography–Tissot’s Indicatrix, specific to sphere-to-plane maps– visualizes distortion using ellipses. Inspired by this idea, we describe the hypertrix: a method for representing distortions that occur when data is projected from arbitrarily high dimensions onto a 2D plane. We demonstrate our technique through synthetic and real-world datasets, and describe how this indicatrix can guide interpretations of nonlinear dimensionality reduction.