Deterministic convergence of an online gradient method for BP neural networks

Wei Wu, Guorui Feng, Zhengxue Li, Yuesheng XuIEEE Trans. Neural Networks 2005 · OpenReview

Abstract

Online gradient methods are widely used for training feedforward neural networks. We prove in this paper a convergence theorem for an online gradient method with variable step size for backward propagation (BP) neural networks with a hidden layer. Unlike most of the convergence results that are of probabilistic and nonmonotone nature, the convergence result that we establish here has a deterministic and monotone nature.
BibTeX
@article{wu2005deterministic,
  author = {Wei Wu and Guorui Feng and Zhengxue Li and Yuesheng Xu},
  title = {{Deterministic convergence of an online gradient method for BP neural networks}},
  journal = {IEEE Trans. Neural Networks},
  year = {2005},
  url = {https://openreview.net/forum?id=9J1TLgca4Rv},
}

Loading related content...

Figure