The local geometry of finite mixtures

Abstract

We establish that for q>=1, the class of convex combinations of q translates of a smooth probability density has local doubling dimension proportional to q. The key difficulty in the proof is to control the local geometric structure of mixture classes. Our local geometry theorem yields a bound on the (bracketing) metric entropy of a class of normalized densities, from which a local entropy bound is deduced by a general slicing procedure.

25 pages

Cited by in corpus (4)

• About the posterior distribution in hidden Markov models with unknown number of states
• Borrowing strengh in hierarchical Bayes: Posterior concentration of the Dirichlet base measure
• Fundamental limits for learning hidden Markov model parameters
• Optimal estimation of high-dimensional location Gaussian mixtures