Neural network trained with spatial errors
US7409372B2 · kind B2 · utility
19Cited by
21References
37Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Jun 20, 2003 |
| Grant date | Aug 5, 2008 |
| Priority date | — |
| Expiry date | Aug 5, 2025 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N3/09
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A neural network is trained with input data. The neural network is used to rescale the input data. Errors for the rescaled values are determined, and neighborhoods of the errors are used adjust connection weights of the neural network.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.