Deep Gaussian Mixture Models
This work proposes a novel modeling approach for data analysis, but it appears incremental as it builds on existing Gaussian mixture and deep learning concepts without demonstrating broad applications or specific gains.
The paper introduces Deep Gaussian Mixture Models (DGMMs), a hierarchical network of latent variables with Gaussian mixtures at each layer, to flexibly model complex data relationships, and addresses overparameterization by incorporating dimension reduction through factor models.
Deep learning is a hierarchical inference method formed by subsequent multiple layers of learning able to more efficiently describe complex relationships. In this work, Deep Gaussian Mixture Models are introduced and discussed. A Deep Gaussian Mixture model (DGMM) is a network of multiple layers of latent variables, where, at each layer, the variables follow a mixture of Gaussian distributions. Thus, the deep mixture model consists of a set of nested mixtures of linear models, which globally provide a nonlinear model able to describe the data in a very flexible way. In order to avoid overparameterized solutions, dimension reduction by factor models can be applied at each layer of the architecture thus resulting in deep mixtures of factor analysers.