Deep AutoRegressive Networks
This work addresses the challenge of efficient sampling and representation learning in generative models, with incremental improvements in performance.
The paper tackles the problem of learning hierarchical distributed representations from data by introducing a deep generative autoencoder with autoregressive connections, achieving state-of-the-art generative performance on datasets like MNIST and Atari 2600 games.
We introduce a deep, generative autoencoder capable of learning hierarchies of distributed representations from data. Successive deep stochastic hidden layers are equipped with autoregressive connections, which enable the model to be sampled from quickly and exactly via ancestral sampling. We derive an efficient approximate parameter estimation method based on the minimum description length (MDL) principle, which can be seen as maximising a variational lower bound on the log-likelihood, with a feedforward neural network implementing approximate inference. We demonstrate state-of-the-art generative performance on a number of classic data sets: several UCI data sets, MNIST and Atari 2600 games.