Training a Subsampling Mechanism in Expectation
Abstract
We describe a mechanism for subsampling sequences and show how to compute its expected output so that it can be trained with standard backpropagation. We test this approach on a simple toy problem and discuss its shortcomings.
Cite
Text
Raffel and Lawson. "Training a Subsampling Mechanism in Expectation." International Conference on Learning Representations, 2017.Markdown
[Raffel and Lawson. "Training a Subsampling Mechanism in Expectation." International Conference on Learning Representations, 2017.](https://mlanthology.org/iclr/2017/raffel2017iclr-training/)BibTeX
@inproceedings{raffel2017iclr-training,
title = {{Training a Subsampling Mechanism in Expectation}},
author = {Raffel, Colin and Lawson, Dieterich},
booktitle = {International Conference on Learning Representations},
year = {2017},
url = {https://mlanthology.org/iclr/2017/raffel2017iclr-training/}
}