Abstract | ||
---|---|---|
Standard error back-propagation requires output data that is scaled to lie within the active area of the activation function. We show that normalizing data to conform to this requirement is not only a time-consuming process, but can also introduce inaccuracies in modelling of the data. In this paper we propose the gamma learning rule for feedforward neural networks which eliminates the need to scale output data before training. We show that the utilization of "self-scaling" units results in... |
Year | DOI | Venue |
---|---|---|
1995 | 10.1007/3-540-59497-3_198 | IWANN |
Keywords | Field | DocType |
gamma learning,automatic scaling,feedforward neural networks,activation function,feedforward neural network,back propagation,standard error | Convergence (routing),Feedforward neural network,Normalization (statistics),Pattern recognition,Computer science,Activation function,Recurrent neural network,Probabilistic neural network,Learning rule,Time delay neural network,Artificial intelligence,Machine learning | Conference |
Volume | ISSN | ISBN |
930 | 0302-9743 | 3-540-59497-3 |
Citations | PageRank | References |
3 | 0.43 | 3 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andries Petrus Engelbrecht | 1 | 2183 | 125.32 |
Ian Cloete | 2 | 132 | 16.61 |
J. Geldenhuys | 3 | 3 | 0.43 |
Jacek M. Zurada | 4 | 2553 | 226.22 |