A Greedy EM Algorithm for Gaussian Mixture Learning

By Nikos A. Vlassis, Aristidis C. Likas. Neural Processing Letters, 2002

340 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/07be056474e34afad7ed10e5c743ebe405c155a7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Benedikt Pfülb
2. Alexander Rainer Tassilo Gepperth
3. Jakob Verbeek
4. Ben J. A. Kröse
5. Nikos A. Vlassis
6. Daniel Cohen-Or
7. Mohamed S. Kamel
8. F. Karray

In Signal Processing

1. Scott Makeig
2. Arnaud Delorme
3. Jakob Verbeek
4. Christian Andreas Kothe
5. Ben J. A. Kröse
6. Nikos A. Vlassis
7. Daniel Cohen-Or
8. Mohamed S. Kamel

Citations in the corpus

• Locality-Preserving Dimensionality Reduction and Classification for Hyperspectral Image Analysis (2012, methodology): “Once the optimal number of components (K) per GMM has been determined, the parameters for the mixture model can be estimated by the expectation–maximization (EM) algorithm [21], an iterative optimization strategy.”
• Gradient-based training of Gaussian Mixture Models in High-Dimensional Spaces (2019, background): “There are several proposals for what is termed “incremental” or “online” EM for training GMMs Vlassis & Likas (2002); Engel & Heinen (2010); Pinto & Engel (2015); Cederborg et al. (2010); Song & Wang (2005); Kristan et al. (2008), allowing GMMs to be trained by providing one sample at a time.”
• Consistent mesh partitioning and skeletonisation using the shape diameter function (2008, methodology): “This is achieved using the expectationmaximization (EM) algorithm, which takes very little time since the histogram is only 1D. For further details regarding GMM and EM, we refer the reader to [7, 33 ].”
• The global k-means clustering algorithm (2003, methodology): “A similar ‘trick’ has been used in the re-lated problems of greedy mixture density estimation using the EM algorithm [4] and principal curve 2tting [5].”
• Survey on speech emotion recognition: Features, classification schemes, and databases (2011, methodology): “Gaussian mixture model is a probabilistic model for density estimation using a convex combination of multi-variate normal densities [133].”
• Efficient Greedy Learning of Gaussian Mixture Models (2003, methodology): “Vlassis and Likas (2002) proposed using n candidate components.”
• EEGLAB, SIFT, NFT, BCILAB, and ERICA: New Tools for Advanced EEG Processing (2011, methodology): “(1) Time-frequency window selection (2) Frequency-domain DAL (FDAL) (vi) Gaussian mixture models (GMM), 9 methods [35–37])”