Role of biases in neural network models
The capacity problem for multi-layer networks has proven especially elusive. Our calculation of the capacity of multi-layer networks built by constructive algorithms relies heavily on the existence of biases in the basic building block, the binary perceptron. It is the first time where the capacity...
| Main Author: | |
|---|---|
| Published: |
University of Edinburgh
1997
|
| Subjects: | |
| Online Access: | http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.663657 |