Function Approximation with the Sweeping Hinge Algorithm

Don R. Hush, Fernando Lozano, Bill G. Horne

NeurIPS 1997 pp. 535-541

/neurips/1997/hush1997neurips-function/

Abstract

We present a computationally efficient algorithm for function ap(cid:173) proximation with piecewise linear sigmoidal nodes. A one hidden layer network is constructed one node at a time using the method of fitting the residual. The task of fitting individual nodes is accom(cid:173) plished using a new algorithm that searchs for the best fit by solving a sequence of Quadratic Programming problems. This approach of(cid:173) fers significant advantages over derivative-based search algorithms (e.g. backpropagation and its extensions). Unique characteristics of this algorithm include: finite step convergence, a simple stop(cid:173) ping criterion, a deterministic methodology for seeking "good" local minima, good scaling properties and a robust numerical implemen(cid:173) tation.

PDF NeurIPS Semantic Scholar

Cite

Text

Hush et al. "Function Approximation with the Sweeping Hinge Algorithm." Neural Information Processing Systems, 1997.

Markdown

[Hush et al. "Function Approximation with the Sweeping Hinge Algorithm." Neural Information Processing Systems, 1997.](https://mlanthology.org/neurips/1997/hush1997neurips-function/)

BibTeX

@inproceedings{hush1997neurips-function,
  title     = {{Function Approximation with the Sweeping Hinge Algorithm}},
  author    = {Hush, Don R. and Lozano, Fernando and Horne, Bill G.},
  booktitle = {Neural Information Processing Systems},
  year      = {1997},
  pages     = {535-541},
  url       = {https://mlanthology.org/neurips/1997/hush1997neurips-function/}
}