The Learning Dynamcis of a Universal Approximator

Abstract

The learning properties of a universal approximator, a normalized committee machine with adjustable biases, are studied for on-line back-propagation learning. Within a statistical mechanics frame(cid:173) work, numerical studies show that this model has features which do not exist in previously studied two-layer network models with(cid:173) out adjustable biases, e.g., attractive suboptimal symmetric phases even for realizable cases and noiseless data.

Cite

Text

West et al. "The Learning Dynamcis of a Universal Approximator." Neural Information Processing Systems, 1996.

Markdown

[West et al. "The Learning Dynamcis of a Universal Approximator." Neural Information Processing Systems, 1996.](https://mlanthology.org/neurips/1996/west1996neurips-learning/)

BibTeX

@inproceedings{west1996neurips-learning,
  title     = {{The Learning Dynamcis of a Universal Approximator}},
  author    = {West, Ansgar H. L. and Saad, David and Nabney, Ian T.},
  booktitle = {Neural Information Processing Systems},
  year      = {1996},
  pages     = {288-294},
  url       = {https://mlanthology.org/neurips/1996/west1996neurips-learning/}
}