NIPS Proceedingsβ

RNADE: The real-valued neural autoregressive density-estimator

Part of: Advances in Neural Information Processing Systems 26 (NIPS 2013)

[PDF] [BibTeX] [Supplemental] [Reviews]


Conference Event Type: Poster


We introduce RNADE, a new model for joint density estimation of real-valued vectors. Our model calculates the density of a datapoint as the product of one-dimensional conditionals modeled using mixture density networks with shared parameters. RNADE learns a distributed representation of the data, while having a tractable expression for the calculation of densities. A tractable likelihood allows direct comparison with other methods and training by standard gradient-based optimizers. We compare the performance of RNADE on several datasets of heterogeneous and perceptual data, finding it outperforms mixture models in all but one case.