English

Do Convolutional Networks need to be Deep for Text Classification ?

Computation and Language 2017-07-14 v1

Abstract

We study in this work the importance of depth in convolutional models for text classification, either when character or word inputs are considered. We show on 5 standard text classification and sentiment analysis tasks that deep models indeed give better performances than shallow networks when the text input is represented as a sequence of characters. However, a simple shallow-and-wide network outperforms deep models such as DenseNet with word inputs. Our shallow word model further establishes new state-of-the-art performances on two datasets: Yelp Binary (95.9\%) and Yelp Full (64.9\%).

Keywords

Cite

@article{arxiv.1707.04108,
  title  = {Do Convolutional Networks need to be Deep for Text Classification ?},
  author = {Hoa T. Le and Christophe Cerisara and Alexandre Denis},
  journal= {arXiv preprint arXiv:1707.04108},
  year   = {2017}
}
R2 v1 2026-06-22T20:45:55.053Z