next up previous contents
Next: Motivation: 4-Bit-Parity Problem Up: Modularity Previous: Modularity Derived from Psychology

Modularity in ANN

The most used artificial neural networks have a monolithic structure. A lot of the models work on fully connected networks or layers (Hopfield or multilayer Perceptron). These networks perform well on a very small input space. However the complexity increases and the performance decreases rapidly with a growing input dimension [koho88]. The next paragraph gives a motivation for modular ANN.




next up previous contents
Next: Motivation: 4-Bit-Parity Problem Up: Modularity Previous: Modularity Derived from Psychology

Albrecht Schmidt
Mit Okt 4 16:45:34 CEST 2000