On a Magnitude Preserving Iterative MAXnet Algorithm
- 1 March 1992
- journal article
- Published by MIT Press in Neural Computation
- Vol. 4 (2) , 224-233
- https://doi.org/10.1162/neco.1992.4.2.224
Abstract
A new iterative maximum picking neural net (MAXnet) is presented. This formulation determines the value and the location either for a unique maximum or for multiple maxima. This new net converges, for many commonly occurring distributions, in O(log M) iterations using only simple computing elements.Keywords
This publication has 3 references indexed in Scilit:
- Minimum distance automata in parallel networks for optimum classificationNeural Networks, 1989
- Neurons with graded response have collective computational properties like those of two-state neurons.Proceedings of the National Academy of Sciences, 1984
- Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural NetworksStudies in Applied Mathematics, 1973