Advanced Search
DSpace@MIT

An Accelerated Chow and Liu Algorithm: Fitting Tree Distributions to High Dimensional Sparse Data

Research and Teaching Output of the MIT Community

Show simple item record

dc.contributor.author Meila, Marina en_US
dc.date.accessioned 2004-10-08T20:37:13Z
dc.date.available 2004-10-08T20:37:13Z
dc.date.issued 1999-01-01 en_US
dc.identifier.other AIM-1652 en_US
dc.identifier.uri http://hdl.handle.net/1721.1/6676
dc.description.abstract Chow and Liu introduced an algorithm for fitting a multivariate distribution with a tree (i.e. a density model that assumes that there are only pairwise dependencies between variables) and that the graph of these dependencies is a spanning tree. The original algorithm is quadratic in the dimesion of the domain, and linear in the number of data points that define the target distribution $P$. This paper shows that for sparse, discrete data, fitting a tree distribution can be done in time and memory that is jointly subquadratic in the number of variables and the size of the data set. The new algorithm, called the acCL algorithm, takes advantage of the sparsity of the data to accelerate the computation of pairwise marginals and the sorting of the resulting mutual informations, achieving speed ups of up to 2-3 orders of magnitude in the experiments. en_US
dc.format.extent 1375477 bytes
dc.format.extent 434859 bytes
dc.format.mimetype application/postscript
dc.format.mimetype application/pdf
dc.language.iso en_US
dc.relation.ispartofseries AIM-1652 en_US
dc.title An Accelerated Chow and Liu Algorithm: Fitting Tree Distributions to High Dimensional Sparse Data en_US


Files in this item

Name Size Format Description
AIM-1652.ps 1.311Mb Postscript
AIM-1652.pdf 424.6Kb PDF

This item appears in the following Collection(s)

Show simple item record

MIT-Mirage