Learning Diagonal Gaussian Mixture Models and Incomplete Tensor Decompositions

2021/02/08 by Bingni Guo, Guo, Bingni, Jiawang Nie +3 · 2 citations
Computer Science · Mathematics · #FOS: Mathematics #Machine Learning and Algorithms #Numerical Analysis (math.NA) #Tensor decomposition and applications

paper · pdf · doi:10.48550/arxiv.2102.04500

openalex publication_date 2021/02/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper studies how to learn parameters in diagonal Gaussian mixture models. The problem can be formulated as computing incomplete symmetric tensor decompositions. We use generating polynomials to compute incomplete symmetric tensor decompositions and approximations. Then the tensor approximation method is used to learn diagonal Gaussian mixture models. We also do the stability analysis. When the first and third order moments are sufficiently accurate, we show that the obtained parameters for the Gaussian mixture models are also highly accurate. Numerical experiments are also provided.

Citations

• Most Tensor Problems Are NP-Hard

Cited by

• Learning Polynomial Activation Functions for Deep Neural Networks
• Diagonally-Weighted Generalized Method of Moments Estimation for Gaussian Mixture Modeling

Related

• Diagonal Gaussian Mixture Models and Higher Order Tensor Decompositions
• Tensor Moments of Gaussian Mixture Models: Theory and Applications
• Overcomplete order-3 tensor decomposition, blind deconvolution and Gaussian mixture models
• Moment Estimation for Nonparametric Mixture Models Through Implicit Tensor Decomposition
• Learning Mixtures of Gaussians in High Dimensions