• DOI:10.1109/ICASSP.2016.7472159
Corpus ID: 516518

Batch normalized recurrent neural networks

@article{Laurent2016BatchNR,
  title={Batch normalized recurrent neural networks},
  author={C{\'e}sar Laurent and G. Pereyra and Philemon Brakel and Y. Zhang and Yoshua Bengio},
  journal={2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2016},
  pages={2657-2661}
}

César Laurent, G. Pereyra, +2 authors Yoshua Bengio

Published 2016

Computer Science, Mathematics

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Recurrent Neural Networks (RNNs) are powerful models for sequential data that have the potential to learn long-term dependencies. However, they are computationally expensive to train and difficult to parallelize. Recent work has shown that normalizing intermediate representations of neural networks can significantly improve convergence rates in feed-forward neural networks [1]. In particular, batch normalization, which uses mini-batch statistics to standardize features, was shown to… CONTINUE READING

View on IEEE

arxiv.org

Save to Library

Create Alert

Cite

Launch Research Feed