Iterative Neural Autoregressive Distribution Estimator (NADE-k)
Abstract
Training of the neural autoregressive density estimator (NADE) can be viewed as doing one step of probabilistic inference on missing values in data. We propose a new model that extends this inference scheme to multiple steps, arguing that it is easier to learn to improve a reconstruction in steps rather than to learn to reconstruct in a single inference step. The proposed model is an unsupervised building block for deep learning that combines the desirable properties of NADE and multi-predictive training: (1) Its test likelihood can be computed analytically, (2) it is easy to generate independent samples from it, and (3) it uses an inference engine that is a superset of variational inference for Boltzmann machines. The proposed NADE-k is competitive with the state-of-the-art in density estimation on the two datasets tested.
Cite
@article{arxiv.1406.1485,
title = {Iterative Neural Autoregressive Distribution Estimator (NADE-k)},
author = {Tapani Raiko and Li Yao and Kyunghyun Cho and Yoshua Bengio},
journal= {arXiv preprint arXiv:1406.1485},
year = {2014}
}
Comments
Accepted at Neural Information Processing Systems (NIPS) 2014