Approximability of Probability Distributions

@inproceedings{Beygelzimer2003ApproximabilityOP,
  title={Approximability of Probability Distributions},
  author={Alina Beygelzimer and Irina Rish},
  booktitle={NIPS},
  year={2003}
}
We consider the question of how well a given distribution can be approximated with probabilistic graphical models. We introduce a new parameter,effective treewidth, that captures the degree of approximability as a tradeoff between the accuracy and the complexity of approx imation. We present a simple approach to analyzing achievable tradeo ffs that exploits the threshold behavior of monotone graph properties , and provide experimental results that support the approach. 

From This Paper

Figures, tables, and topics from this paper.

References

Publications referenced by this paper.
Showing 1-10 of 12 references

Every monotone graph property has a sh arp t reshold.Proceedings of the American

  • E. Friedgut, G. Kalai
  • Mathematical Society,
  • 1996
Highly Influential
5 Excerpts

Reasoning in Intelligent Systems: Networks of Plausible Inference

  • J. Pearl.Probabilistic
  • 1988
Highly Influential
5 Excerpts

Similar Papers

Loading similar papers…