Abstract
A least square approximation method is proposed to reduce the number of hidden units of a trained multilayer perceptron artificial neural network structure. In this method, hidden neurons contributing the most to the net function of the output layer will be retained while the hidden units contributing the least will be removed. It is shown theoretically that the proposed method minimizes the Frobenius norm of the approximation error, hence the name Frobenius approximation reduction method (FARM). Also reported are simulation results on ECG (electrocardiogram) classifications. The results support the theoretical predictions and yield very encouraging performances.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings. IJCNN - International Joint Conference on Neural Networks |
| Editors | Anon |
| Publisher | Publ by IEEE |
| Pages | 163-168 |
| Number of pages | 6 |
| ISBN (Print) | 0780301641 |
| State | Published - Jan 1 1992 |
| Event | International Joint Conference on Neural Networks - IJCNN-91-Seattle - Seattle, WA, USA Duration: Jul 8 1991 → Jul 12 1991 |
Other
| Other | International Joint Conference on Neural Networks - IJCNN-91-Seattle |
|---|---|
| City | Seattle, WA, USA |
| Period | 7/8/91 → 7/12/91 |
All Science Journal Classification (ASJC) codes
- Engineering(all)