paper-with-me

홈 › Papers

Local Minima Structures in Gaussian Mixture Models

2020-09-28 · Yudong Chen, Dogyoon Song, Xumei Xi, Yuqian Zhang

We investigate the landscape of the negative log-likelihood function of Gaussian Mixture Models (GMMs) with a general number of components in the population limit. As the objective function is non-convex, there can be multiple local minima that are not globally optimal, even for well-separated mixture models. Our study reveals that all local minima share a common structure that partially identifies the cluster centers (i.e., means of the Gaussian components) of the true location mixture. Specifically, each local minimum can be represented as a non-overlapping combination of two types of sub-configurations: fitting a single mean estimate to multiple Gaussian components or fitting multiple estimates to a single true component. These results apply to settings where the true mixture components satisfy a certain separation condition, and are valid even when the number of components is over- or under-specified. We also present a more fine-grained analysis for the setting of one-dimensional GMMs with three components, which provide sharper approximation error bounds with improved dependence on the separation.

📄 PDF Abstract BibTeX arXiv:2009.13040

Code (0)

등록된 구현이 없습니다.

Tasks

valid

Similar Papers 제목 키워드 기반

Minimax-Optimal Dimension-Reduced Clustering for High-Dimensional Nonspherical Mixtures

2025-02-04 · Chengzhu Huang, Yuqi Gu

In mixture models, nonspherical (anisotropic) noise within each cluster is widely present in real-world data. We study both the minimax rate and optimal statistical procedure for clustering under high-dimensional nonsphe…

ClusteringDimensionality Reduction

Robust scalable initialization for Bayesian variational inference with multi-modal Laplace approximations

2023-07-12 · Wyatt Bridgman, Reese Jones, Mohammad Khalil

For predictive modeling relying on Bayesian inversion, fully independent, or ``mean-field'', Gaussian distributions are often used as approximate probability density functions in variational inference since the number of…

Variational Inference

GOGMA: Globally-Optimal Gaussian Mixture Alignment

2016-03-01 · CVPR 2016 6 · Dylan Campbell, Lars Petersson

Gaussian mixture alignment is a family of approaches that are frequently used for robustly solving the point-set registration problem. However, since they use local optimisation, they are susceptible to local minima and …

Comparing EM with GD in Mixture Models of Two Components

2019-07-08 · Guojun Zhang, Pascal Poupart, George Trimponias

The expectation-maximization (EM) algorithm has been widely used in minimizing the negative log likelihood (also known as cross entropy) of mixture models. However, little is understood about the goodness of the fixed po…

Vocal Bursts Valence Prediction

Structures of Spurious Local Minima in $k$-means

2020-02-16 · Wei Qian, Yuqian Zhang, Yudong Chen

$k$-means clustering is a fundamental problem in unsupervised learning. The problem concerns finding a partition of the data points into $k$ clusters such that the within-cluster variation is minimized. Despite its impor…

Clustering