Maaten, L. van den: Accelerating t-SNE using Tree-Based Algorithms (2014)
0.02
0.016414853 = product of:
0.08207426 = sum of:
0.08207426 = weight(_text_:o in 3886) [ClassicSimilarity], result of:
0.08207426 = score(doc=3886,freq=2.0), product of:
0.21151194 = queryWeight, product of:
5.017288 = idf(docFreq=795, maxDocs=44218)
0.042156626 = queryNorm
0.38803607 = fieldWeight in 3886, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
5.017288 = idf(docFreq=795, maxDocs=44218)
0.0546875 = fieldNorm(doc=3886)
0.2 = coord(1/5)
- Abstract
- The paper investigates the acceleration of t-SNE-an embedding technique that is commonly used for the visualization of high-dimensional data in scatter plots-using two tree-based algorithms. In particular, the paper develops variants of the Barnes-Hut algorithm and of the dual-tree algorithm that approximate the gradient used for learning t-SNE embeddings in O(N*logN). Our experiments show that the resulting algorithms substantially accelerate t-SNE, and that they make it possible to learn embeddings of data sets with millions of objects. Somewhat counterintuitively, the Barnes-Hut variant of t-SNE appears to outperform the dual-tree variant.