# Mixtures of Multivariate Power Exponential Distributions

arXiv: https://arxiv.org/abs/1506.04137 · HTML: https://mlsanity.com/paper/1506.04137/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2015.

## Built on by (most cited first, 6 of 6)

- [Vine copula mixture models and clustering for non-Gaussian data](https://mlsanity.com/paper/2102.03257/) (2021)
- [Adaptive Fuzzy C-Means with Graph Embedding](https://mlsanity.com/paper/2405.13427/) (2024)
- [Flat-topped Probability Density Functions for Mixture Models](https://mlsanity.com/paper/2203.17027/) (2022)
- [Finite mixture of skewed sub-Gaussian stable distributions](https://mlsanity.com/paper/2205.14067/) (2022)
- [Clustering Three-Way Data with Outliers](https://mlsanity.com/paper/2310.05288/) (2023)
- [Finite Mixtures of Multivariate Poisson-Log Normal Factor Analyzers for Clustering Count Data](https://mlsanity.com/paper/2311.07762/) (2023)

## Abstract (arXiv)

An expanded family of mixtures of multivariate power exponential distributions is introduced. While fitting heavy-tails and skewness has received much attention in the model-based clustering literature recently, we investigate the use of a distribution that can deal with both varying tail-weight and peakedness of data. A family of parsimonious models is proposed using an eigen-decomposition of the scale matrix. A generalized expectation-maximization algorithm is presented that combines convex optimization via a minorization-maximization approach and optimization based on accelerated line search algorithms on the Stiefel manifold. Lastly, the utility of this family of models is illustrated using both toy and benchmark data.