Classifier Weighted Mixture models
This is an incremental improvement for researchers in machine learning, specifically in density estimation and variational inference.
The paper tackles the limited expressivity of standard mixture models by introducing classifier-weighted mixtures, which replace constant weights with functional weights from a classifier, resulting in enhanced expressivity for variational estimation without increasing model complexity.
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.