Learning when to skim and when to read
Computation and Language
2017-12-18 v1
Abstract
Many recent advances in deep learning for natural language processing have come at increasing computational cost, but the power of these state-of-the-art models is not needed for every example in a dataset. We demonstrate two approaches to reducing unnecessary computation in cases where a fast but weak baseline classier and a stronger, slower model are both available. Applying an AUC-based metric to the task of sentiment classification, we find significant efficiency gains with both a probability-threshold method for reducing computational cost and one that uses a secondary decision network.
Cite
@article{arxiv.1712.05483,
title = {Learning when to skim and when to read},
author = {Alexander Rosenberg Johansen and Richard Socher},
journal= {arXiv preprint arXiv:1712.05483},
year = {2017}
}
Comments
8 pages (4 article, 1 references, 3 appendix), 11 figures, 3 tables, published at ACL2017 workshop Repl4NLP