Daphne Koller

Papers

Daphne Koller Publications

Probabilistic Abstraction Hierarchies (2002)

by E. Segal, D. Koller, and D. Ormoneit

Abstract: Many domains are naturally organized in an abstraction hierarchy or taxonomy, where the instances in "nearby" classes in the taxonomy are similar. In this paper, we provide a general probabilistic framework for clustering data into a set of classes organized as a taxonomy, where each class is associated with a probabilistic model from which the data was generated. The clustering algorithm simultaneously optimizes three things: the assignment of data instances to clusters, the models associated with the clusters, and the structure of the abstraction hierarchy. A unique feature of our approach is that it utilizes global optimization algorithms for both of the last two steps, reducing the sensitivity to noise and the propensity to local maxima that are characteristic of algorithms that only take local steps. We provide a theoretical analysis for our algorithm, showing that it converges to a local maximum of the probability of model and data. We present experimental results on synthetic data, and on real data in the domains of gene expression and text.

Download Information

E. Segal, D. Koller, and D. Ormoneit (2002). "Probabilistic Abstraction Hierarchies." Advances in Neural Information Processing Systems (NIPS 2001) (pp. 913-920).

Bibtex citation

@inproceedings{Segal+al:NIPS01,
  title = {Probabilistic Abstraction Hierarchies},
  author = {E. Segal and D. Koller and D. Ormoneit},
  booktitle = {Advances in Neural Information Processing Systems (NIPS 2001)}, 
  address = {Vancouver, Canada}, 
  month = {December},
  year = 2002,
  pages = {913--920},
}

full list