English

Path-Normalized Optimization of Recurrent Neural Networks with ReLU Activations

Machine Learning 2016-05-24 v1 Neural and Evolutionary Computing

Abstract

We investigate the parameter-space geometry of recurrent neural networks (RNNs), and develop an adaptation of path-SGD optimization method, attuned to this geometry, that can learn plain RNNs with ReLU activations. On several datasets that require capturing long-term dependency structure, we show that path-SGD can significantly improve trainability of ReLU RNNs compared to RNNs trained with SGD, even with various recently suggested initialization schemes.

Keywords

Cite

@article{arxiv.1605.07154,
  title  = {Path-Normalized Optimization of Recurrent Neural Networks with ReLU Activations},
  author = {Behnam Neyshabur and Yuhuai Wu and Ruslan Salakhutdinov and Nathan Srebro},
  journal= {arXiv preprint arXiv:1605.07154},
  year   = {2016}
}

Comments

15 pages

R2 v1 2026-06-22T14:07:33.341Z