Classifier Weighted Mixture models

TL;DR

This paper introduces Classifier Weighted Mixture models, which replace fixed mixture weights with classifier-based functions, improving flexibility and sampling without added complexity.

Contribution

It presents a novel extension of mixture models that uses classifiers to define mixture weights, enhancing expressivity and sampling efficiency.

Findings

Enables straightforward density evaluation

Allows explicit sampling

Improves expressivity in variational estimation

Abstract

This paper proposes an extension of standard mixture stochastic models, by replacing the constant mixture weights with functional weights defined using a classifier. Classifier Weighted Mixtures enable straightforward density evaluation, explicit sampling, and enhanced expressivity in variational estimation problems, without increasing the number of components nor the complexity of the mixture components.