An Optimization Method of Layered Neural Networks Based on the Modified Information Criterion

Abstract

This paper proposes a practical optimization method for layered neural networks, by which the optimal model and parameter can be found simultaneously. 'i\Te modify the conventional information criterion into a differentiable function of parameters, and then, min(cid:173) imize it, while controlling it back to the ordinary form. Effective(cid:173) ness of this method is discussed theoretically and experimentally.

Cite

Text

Watanabe. "An Optimization Method of Layered Neural Networks Based on the Modified Information Criterion." Neural Information Processing Systems, 1993.

Markdown

[Watanabe. "An Optimization Method of Layered Neural Networks Based on the Modified Information Criterion." Neural Information Processing Systems, 1993.](https://mlanthology.org/neurips/1993/watanabe1993neurips-optimization/)

BibTeX

@inproceedings{watanabe1993neurips-optimization,
  title     = {{An Optimization Method of Layered Neural Networks Based on the Modified Information Criterion}},
  author    = {Watanabe, Sumio},
  booktitle = {Neural Information Processing Systems},
  year      = {1993},
  pages     = {293-300},
  url       = {https://mlanthology.org/neurips/1993/watanabe1993neurips-optimization/}
}