BRUNO: A Deep Recurrent Model for Exchangeable Data
Iryna Korshunova, Jonas Degrave, Ferenc Huszar, Yarin Gal, Arthur Gretton, Joni Dambre
Abstract
We present a novel model architecture which leverages deep learning tools to perform exact Bayesian inference on sets of high dimensional, complex observations. Our model is provably exchangeable, meaning that the joint distribution over observations is invariant under permutation: this property lies at the heart of Bayesian inference. The model does not require variational approximations to train, and new samples can be generated conditional on previous samples, with cost linear in the size of the conditioning set. The advantages of our architecture are demonstrated on learning tasks that require generalisation from short observed sequences while modelling sequence variability, such as conditional image generation, few-shot learning, and anomaly detection.
BibTeX
@inproceedings{NEURIPS2018_1b9f3826,
author = {Korshunova, Iryna and Degrave, Jonas and Huszar, Ferenc and Gal, Yarin and Gretton, Arthur and Dambre, Joni},
booktitle = {Advances in Neural Information Processing Systems},
editor = {S. Bengio and H. Wallach and H. Larochelle and K. Grauman and N. Cesa-Bianchi and R. Garnett},
pages = {},
publisher = {Curran Associates, Inc.},
title = {BRUNO: A Deep Recurrent Model for Exchangeable Data},
url = {https://proceedings.neurips.cc/paper_files/paper/2018/file/1b9f38268c50805669fd8caf8f3cc84a-Paper.pdf},
volume = {31},
year = {2018}
}