Asia-Pacific Journal of Neural Networks and Its Applications
| Volume 1, No. 1, 2017, pp 7-14 | ||
| Abstract |
The Characteristic Function of CoreNet (Multi-Level Single-Layer Artificial Neural Networks)
|
|
Abstract
The interpretation of hidden layer is an unsolved problem in Neural Networks. This paper suggested a characteristic function of CoreNet which has an input(p levels) and an output(q levels) with single-layer as a basic circuit of neural networks. I have suggested an equation, 𝑡 2 − 𝑝(𝑞 − 1)𝑡 + 1 2 (𝑝 − 1)(𝑞 − 1)(𝑝𝑞 − 2𝑝 + 2), which depict a CoreNet(p input levels and q output levels). It has also been shown that some of the problems with hidden layers can be solved by using the CoreNet