Abstract
A new neural model which has a Gaussian activation function is presented. This model is referred to as the Gaussian perceptron. For the training of single-layered Gaussian perceptrons, the Guassian perceptron learning algorithm, which is a variant of the conventional perceptron learning algorithm, is presented. The winner-take-all algorithm is proposed as a multilayer training algorithm. A number of examples are presented along with the comparison with backpropagation networks, which demonstrate the performance of Gaussian perceptron networks.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1593-1598 |
| Number of pages | 6 |
| Journal | Proceedings of the IEEE International Conference on Systems, Man and Cybernetics |
| Volume | 3 |
| State | Published - Dec 1 1991 |
| Event | Conference Proceedings of the 1991 IEEE International Conference on Systems, Man, and Cybernetics - Charlottesville, VA, USA Duration: Oct 13 1991 → Oct 16 1991 |