Variational Inference of Finite Generalized Gaussian Mixture Models

被引：0

作者：

Amudala, Srikanth ^{[1
]}

Ali, Samr ^{[2
]}

Najar, Fatma ^{[1
]}

Bouguila, Nizar ^{[1
]}

机构：

[1] Concordia Univ, Concordia Inst Informat Syst Engn, Montreal, PQ, Canada

[2] Concordia Univ, Dept Elect & Comp Engn, Montreal, PQ, Canada

来源：

2019 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2019) | 2019年

关键词：

generalized gaussian mixture model; variational inference; taylor approximation; binary classification; image segmentation; SUBBAND IMAGE DATA; DENSITY; CLASSIFICATION; DISTRIBUTIONS; PARAMETER; SHAPE; DCT;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a variational learning framework to analyze finite g eneralized G aussian m ixture models (GGMM). The model incorporates several mixtures that are widely used in signal and image processing applications. The motivation behind this work is the shape flexibility characteristics of the generalized Gaussian distribution (GGD) because of which it can be applied to different types of data. We present a method to evaluate the posterior distribution and Bayes estimators using the variational expectation-maximization algorithm. The effective number of components of the GGMM is determined automatically. The test results show the adequacy of the proposed algorithm by applying it to medical, astrological, and image segmentation applications; while comparing it with various other approaches.

引用

页码：2433 / 2439

页数：7

共 50 条

[1] Variational Inference of Finite Asymmetric Gaussian Mixture Models
Song, Ziyang
Bregu, Ornela
Ali, Samr
Bouguila, Nizar
[J]. 2019 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2019), 2019, : 2448 - 2454
[2] Variational Inference of Infinite Generalized Gaussian Mixture Models with Feature Selection
Amudala, Srikanth
Ali, Samr
Bouguila, Nizar
[J]. 2020 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), 2020, : 120 - 127
[3] Sparse Variational Inference for Generalized Gaussian Process Models
Sheth, Rishit
Wang, Yuyang
Khardon, Roni
[J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 37, 2015, 37 : 1302 - 1311
[4] Trust-region variational inference with gaussian mixture models
Arenz, Oleg
Zhong, Mingjun
Neumann, Gerhard
[J]. Journal of Machine Learning Research, 2020, 21
[5] Trust-Region Variational Inference with Gaussian Mixture Models
Arenz, Oleg
Zhong, Mingjun
Neumann, Gerhard
[J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2020, 21
[6] Gaussian Variational Approximate Inference for Generalized Linear Mixed Models
Ormerod, J. T.
Wand, M. P.
[J]. JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2012, 21 (01) : 2 - 17
[7] Variational inference and sparsity in high-dimensional deep Gaussian mixture models
Kock, Lucas
Klein, Nadja
Nott, David J.
[J]. STATISTICS AND COMPUTING, 2022, 32 (05)
[8] Variational inference and sparsity in high-dimensional deep Gaussian mixture models
Lucas Kock
Nadja Klein
David J. Nott
[J]. Statistics and Computing, 2022, 32
[9] Bayesian learning of finite generalized Gaussian mixture models on images
Elguebaly, Tarek
Bouguila, Nizar
[J]. SIGNAL PROCESSING, 2011, 91 (04) : 801 - 820
[10] Variational learning for Gaussian mixture models
Nasios, Nikolaos
Bors, Adrian G.
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2006, 36 (04): : 849 - 862

← 1 2 3 4 5 →