Predicting Parameters in Deep Learning
Machine Learning
2014-10-28 v2 Neural and Evolutionary Computing
Machine Learning
Abstract
We demonstrate that there is significant redundancy in the parameterization of several deep learning models. Given only a few weight values for each feature it is possible to accurately predict the remaining values. Moreover, we show that not only can the parameter values be predicted, but many of them need not be learned at all. We train several different architectures by learning only a small number of weights and predicting the rest. In the best case we are able to predict more than 95% of the weights of a network without any drop in accuracy.
Cite
@article{arxiv.1306.0543,
title = {Predicting Parameters in Deep Learning},
author = {Misha Denil and Babak Shakibi and Laurent Dinh and Marc'Aurelio Ranzato and Nando de Freitas},
journal= {arXiv preprint arXiv:1306.0543},
year = {2014}
}