The Third Generation of Neural Networks: Beyond simple nodes and linear synapses

The paper discusses an analogue neural network concept in which the neuron is split up into parts (more precise, we consider the dendrite as a seperate and subdivided entity) and the synapse is non-linear. The neural network constructed in this way belongs to the third generation of neural networks. Keywords--- (Artificial) Neural Networks, Neural Computing, Dendritic Computation, Biologicinspired subsystems, Nano-scale electronics, Learning, Compartmental modeling. I. Introduction S INCE the 1990s, neural networks have become one of the tools in the field of information processing that provides the best results both regarding quality of the outcome and the ease of implementation in situations where a model-based or parametric approach is difficult to formulate. Examples of areas where results obtained are arguably better, in some sens, than those expected from various conventional techniques are [1]: nonlinear time-varying control and prediction, telecommunication systems, (medical..