A new implementation of k-MLE for mixture modeling of Wishart distributions
Résumé
We describe an original implementation of k-Maximum Like- lihood Estimator (k-MLE)[1], a fast algorithm for learning finite statisti- cal mixtures of exponential families. Our version converges to a local maximum of the complete likelihood while guaranteeing not to have empty clusters. To initialize k-MLE, we propose a careful and greedy strategy inspired by k-means++ which selects automatically cluster cen- ters and their number. The paper gives all details for using k-MLE with mixtures of Wishart (WMMs). Finally, we propose to use the Cauchy- Schwartz divergence as a comparison measure between two WMMs and give a general methodology for building a motion retrieval system.
Origine : Fichiers produits par l'(les) auteur(s)