Multi-layer neural networks for handwritten digit recognition
This report documents the underlining theories and neural network that lead to the development of handwritten digit recognition architecture that are capable of recognizing handwritten digit with recognition accuracy of up to 76%.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
2008
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/4543 |