AISTATS 2015poster9 citations

Symmetric Iterative Proportional Fitting

Sven Kurras

Abstract

Iterative Proportional Fitting (IPF) generates from an input matrix W a sequence of matrices that converges, under certain conditions, to a specific limit matrix W*. This limit is the relative-entropy nearest solution to W among all matrices of prescribed row marginals r and column marginals c. We prove this known fact by a novel strategy that contributes a pure algorithmic intuition. Then we focus on the symmetric setting: W=W’ and r=c. Since IPF inherently generates non-symmetric matrices, we introduce two symmetrized variants of IPF. We prove convergence for both of them. Further, we give a novel characterization for the existence of W* in terms of expansion properties of the undirected weighted graph represented by W. Finally, we show how our results contribute to recent work in machine learning.

BibTeX
@InProceedings{pmlr-v38-kurras15,
  title = 	 {{Symmetric Iterative Proportional Fitting}},
  author = 	 {Kurras, Sven},
  booktitle = 	 {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics},
  pages = 	 {526--534},
  year = 	 {2015},
  editor = 	 {Lebanon, Guy and Vishwanathan, S. V. N.},
  volume = 	 {38},
  series = 	 {Proceedings of Machine Learning Research},
  address = 	 {San Diego, California, USA},
  month = 	 {09--12 May},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v38/kurras15.pdf},
  url = 	 {https://proceedings.mlr.press/v38/kurras15.html},
  abstract = 	 {Iterative Proportional Fitting (IPF) generates from an input matrix W a sequence of matrices that converges, under certain conditions, to a specific limit matrix W*. This limit is the relative-entropy nearest solution to W among all matrices of prescribed row marginals r and column marginals c. We prove this known fact by a novel strategy that contributes a pure algorithmic intuition. Then we focus on the symmetric setting: W=W’ and r=c. Since IPF inherently generates non-symmetric matrices, we introduce two symmetrized variants of IPF. We prove convergence for both of them. Further, we give a novel characterization for the existence of W* in terms of expansion properties of the undirected weighted graph represented by W. Finally, we show how our results contribute to recent work in machine learning.}
}
Symmetric Iterative Proportional Fitting · AISTATS 2015