Efficient greedy learning of Gaussian mixture models

Abstract : This article concerns the greedy learning of gaussian mixtures. In the greedy approach, mixture components are inserted into the mixture one after the other. We propose a heuristic for searching for the optimal component to insert. In a randomized manner, a set of candidate new components is generated. For each of these candidates, we find the locally optimal new component and insert it into the existing mixture. The resulting algorithm resolves the sensitivity to initialization of state-of-the-art methods, like expectation maximization, and has running time linear in the number of data points and quadratic in the (final) number of mixture components. Due to its greedy nature, the algorithm can be particularly useful when the optimal number of mixture components is unknown. Experimental results comparing the proposed algorithm to other methods on density estimation and texture segmentation are provided.
Document type :
Journal articles
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download


https://hal.inria.fr/inria-00321487
Contributor : Jakob Verbeek <>
Submitted on : Wednesday, February 16, 2011 - 4:15:42 PM
Last modification on : Monday, September 25, 2017 - 10:08:04 AM
Long-term archiving on : Tuesday, May 17, 2011 - 2:28:00 AM

Files

verbeek03neco.pdf
Files produced by the author(s)

Identifiers

Citation

Jakob Verbeek, Nikos Vlassis, Ben Krose. Efficient greedy learning of Gaussian mixture models. Neural Computation, Massachusetts Institute of Technology Press (MIT Press), 2003, 15 (2), pp.469-485. ⟨http://www.mitpressjournals.org/doi/abs/10.1162/089976603762553004⟩. ⟨10.1162/089976603762553004⟩. ⟨inria-00321487⟩

Share

Metrics

Record views

407

Files downloads

1735