Mixture density networks (MDN) are a class of models obtained by combining a conventional neural network with a mixture density model.
A mixture density network is a deep lookahead network designed to produce the probability density function for a multimodal regression problem.
The underlying multimodal model is a Gaussian mixture model, characterized by the mixture of proportions, means, and covariances of the components of the mixture.
The deep lookahead network learns to produce these properties to enable the calculation of the probability density function.
Welcome, my name is Eric Soupet and I am the administrator of the site elodees.com. elodees.com is a state of the art of Artificial Intelligence and aims to be collaborative, you can now offer content such as articles, events, tutorials, ... so don't hesitate !
