Operational speed improvement for neural network
US5271090A · kind A · utility
13Cited by
7References
17Claims
0Family size
Assignee
Inventor
Key dates
| Filing date | Mar 2, 1992 |
| Grant date | Dec 14, 1993 |
| Priority date | — |
| Expiry date | Mar 2, 2012 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N3/048
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Higher operational speed is obtained without sacrificing computational accuracy and reliability in a neural network by interchanging a computationally complex nonlinear function with a similar but less complex nonlinear function in each neuron or computational element after each neuron of the network has been trained by an appropriate training algorithm for the classifying problem addressed by the neural network. In one exemplary embodiment, a hyperbolic tangent function is replaced by a piecewise linear threshold logic function.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.