Graph Drawing by Stochastic Gradient Descent

Zheng JX, Pawar S, Goodman DFM
IEEE Transactions on Visualization and Computer Graphics (2018)
doi: 10.1109/TVCG.2018.2859997
 

Abstract

A popular method of force-directed graph drawing is multidimensional scaling using graph-theoretic distances as input. We present an algorithm to minimize its energy function, known as stress, by using stochastic gradient descent (SGD) to move a single pair of vertices at a time. Our results show that SGD can reach lower stress levels faster and more consistently than majorization, without needing help from a good initialization. We then present various real-world applications to show how the unique properties of SGD make it easier to produce constrained layouts than previous approaches. We also show how SGD can be directly applied within the sparse stress approximation of Ortmann et al. [1], making the algorithm scalable up to large graphs.

Links

Related software

s(gd)2

Graph layout using stochastic gradient descent.

Categories