Error exponents for composite hypothesis testing of Markov forest distributions
Author(s)Tan, Vincent Yan Fu; Anandkumar, Animashree; Willsky, Alan S.
MetadataShow full item record
The problem of composite binary hypothesis testing of Markov forest (or tree) distributions is considered. The worst-case type-II error exponent is derived under the Neyman-Pearson formulation. Under simple null hypothesis, the error exponent is derived in closed-form and is characterized in terms of the so-called bottleneck edge of the forest distribution. The least favorable distribution for detection is shown to be Markov on the second-best max-weight spanning tree with mutual information edge weights. A necessary and sufficient condition to have positive error exponent is derived.
DepartmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science; Massachusetts Institute of Technology. Laboratory for Information and Decision Systems
Proceedings of the IEEE International Symposium on Information Theory Proceedings (ISIT), 2010
Institute of Electrical and Electronics Engineers (IEEE)
Tan, Vincent Y. F., Animashree Anandkumar, and Alan S. Willsky. “Error Exponents for Composite Hypothesis Testing of Markov Forest Distributions.” IEEE International Symposium on Information Theory Proceedings (ISIT), 2010. 1613–1617. © Copyright 2010 IEEE
Final published version