Learning Compact Convolutional Neural Networks with Nested Dropout
Abstract
Recently, nested dropout was proposed as a method for ordering representation units in autoencoders by their information content, without diminishing reconstruction cost. However, it has only been applied to training fully-connected autoencoders in an unsupervised setting. We explore the impact of nested dropout on the convolutional layers in a CNN trained by backpropagation, investigating whether nested dropout can provide a simple and systematic way to determine the optimal representation size with respect to the desired accuracy and desired task and data complexity.
Cite
Text
Finn et al. "Learning Compact Convolutional Neural Networks with Nested Dropout." International Conference on Learning Representations, 2015.Markdown
[Finn et al. "Learning Compact Convolutional Neural Networks with Nested Dropout." International Conference on Learning Representations, 2015.](https://mlanthology.org/iclr/2015/finn2015iclr-learning/)BibTeX
@inproceedings{finn2015iclr-learning,
title = {{Learning Compact Convolutional Neural Networks with Nested Dropout}},
author = {Finn, Chelsea and Hendricks, Lisa Anne and Darrell, Trevor},
booktitle = {International Conference on Learning Representations},
year = {2015},
url = {https://mlanthology.org/iclr/2015/finn2015iclr-learning/}
}