Deep Prior
Machine Learning
2017-12-19 v2 Machine Learning
Abstract
The recent literature on deep learning offers new tools to learn a rich probability distribution over high dimensional data such as images or sounds. In this work we investigate the possibility of learning the prior distribution over neural network parameters using such tools. Our resulting variational Bayes algorithm generalizes well to new tasks, even when very few training examples are provided. Furthermore, this learned prior allows the model to extrapolate correctly far from a given task's training data on a meta-dataset of periodic signals.
Cite
@article{arxiv.1712.05016,
title = {Deep Prior},
author = {Alexandre Lacoste and Thomas Boquet and Negar Rostamzadeh and Boris Oreshkin and Wonchang Chung and David Krueger},
journal= {arXiv preprint arXiv:1712.05016},
year = {2017}
}
Comments
Workshop paper, Accepted at Bayesian Deep Learning workshop, NIPS 2017