trigger

Author: fradav

published-paper-tsne.qmd

@@ -2,6 +2,7 @@
 abstract: |
   We present a new technique called "t-SNE" that visualizes high-dimensional data by giving each datapoint a location in a two or three-dimensional map.  The technique is a variation of Stochastic Neighbor Embedding [@hinton:stochastic] that is much easier to optimize, and produces significantly better visualizations by reducing the tendency to crowd points together in the center of the map. t-SNE is better than existing techniques at creating a single map that reveals structure at many different scales. This is particularly important for high-dimensional data that lie on several different, but related, low-dimensional manifolds, such as images of objects from multiple classes seen from multiple viewpoints. For visualizing the structure of very large data sets, we show how t-SNE can use random walks on neighborhood graphs to allow the implicit structure of all the data to influence the way in which a subset of the data is displayed. We illustrate the performance of t-SNE on a wide variety of data sets and compare it with many other non-parametric visualization techniques, including Sammon mapping, Isomap, and Locally Linear Embedding. The visualization produced by t-SNE are significantly better than those produced by other techniques on almost all of the data sets.
 ---
+
 # Introduction
 
 Visualization of high-dimensional data is an important problem in many different domains, and deals with data of widely varying dimensionality. Cell nuclei that are relevant to breast cancer, for example, are described by approximately 30 variables @street:nuclear, whereas the pixel intensity vectors used to represent images or the word-count vectors used to represent documents typically have thousands of dimensions. Over the last few decades, a variety of techniques for the visualization of such high-dimensional data have been proposed, many of which are reviewed by @ferreira:visual. Important techniques include iconographic displays such as Chernoff faces @chernoff:use, pixel-based techniques @keim:designing, and techniques that represent the dimensions in the data as vertices in a graph @battista:algorithms. Most of these techniques simply provide tools to display more than two data dimensions, and leave the interpretation of the data to the human observer. This severely limits the applicability of these techniques to real-world data sets that contain thousands of high-dimensional datapoints.