Learning Mixtures of Tree Graphical Models

  title={Learning Mixtures of Tree Graphical Models},
  author={Anima Anandkumar and Daniel J. Hsu and Furong Huang and Sham M. Kakade},
We consider unsupervised estimation of mixtures of discret e graphical models, where the class variable is hidden and each mixture componen t can have a potentially different Markov graph structure and parameters ove r th observed variables. We propose a novel method for estimating the mixture compone nts with provable guarantees. Our output is a tree-mixture model which serves as a good approximation to the underlying graphical model mixture. The sampl e and computational requirements for our… CONTINUE READING
Highly Cited
This paper has 40 citations. REVIEW CITATIONS