att_abstract={{For embedding graphs or high dimensional data, stress model is widely
used when edges length, or distances between items, are
specified. However, traditional full stress model is not scalable, due
to the need for all-pairs shortest path calculation.  A number of
fast approximation algorithms were proposed. While they work well for
some graphs, on graphs of intrinsic high dimensions, such as some
non-rigid graphs, the results are less satisfactory. In this paper we
propose a maxent-stress model.  The method uses the principal of
maximal entropy to deal with the extra degrees of freedom. 
We formulate a force-augmented stress
majorization algorithm to solve the maxent-stress model. Numerical
results show that the algorithm can scale to large
graphs, yet does not degrade on
non-rigid graphs. This also has potential applications to scalable algorithms for statistical multidimensional scaling (MDS) with variable distances.

	att_authors={yh573v, eg3218, sn1789},
	att_copyright_notice={{This version of the work is reprinted here with permission of IEEE for your personal use. Not for redistribution. The definitive version was published in 2012 {{, 2012-09-30}}
	att_tags={graph drawing, multi-dimensional scaling, MDS, maxent},
	author={Yifan Hu and Emden Gansner and Stephen North},
	institution={{ IEEE Transactions on Visualization and Computer Graphics.}},
	title={{ Maxent-Stress Model for Graph Layout}},