Comix: Joint Estimation and Lightspeed Comparison of Mixture Models

Abstract : The Kullback-Leibler divergence is a widespread dis-similarity measure between probability density functions , based on the Shannon entropy. Unfortunately, there is no analytic formula available to compute this divergence between mixture models, imposing the use of costly approximation algorithms. In order to reduce the computational burden when a lot of divergence evaluations are needed, we introduce a sub-class of the mixture models where the component parameters are shared between a set of mixtures and the only degree-of-freedom is the vector of weights of each mixture. This sharing allows to design extremely fast versions of existing dis-similarity measures between mixtures. We demonstrate the effectiveness of our approach by evaluating the quality of the ordering produced by our method on a real dataset.
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-01367923
Contributor : Olivier Schwander <>
Submitted on : Saturday, September 17, 2016 - 11:59:01 AM
Last modification on : Wednesday, March 27, 2019 - 4:41:27 PM
Long-term archiving on : Sunday, December 18, 2016 - 3:44:40 PM

File

icassp2016.pdf
Files produced by the author(s)

Identifiers

Citation

Olivier Schwander, Stéphane Marchand-Maillet, Frank Nielsen. Comix: Joint Estimation and Lightspeed Comparison of Mixture Models. ICASSP 2016, 2016, Shanghai, China. ⟨10.1109/ICASSP.2016.7472117⟩. ⟨hal-01367923⟩

Share

Metrics

Record views

659

Files downloads

274