perceptron

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

The perceptron is the earliest kind of artificial neuron. It has a number of inputs and weights for each. In operation the iputs are multipled by the weights and summed to gib and input caruvation. This is then compared with a threshold function, and if the input activation exceeds the threshold, the perceptron fires. A simple percepttron arranged ina. single layer acnnot solve complex problems such as the XOR problem, but multi-layer perceptrons are hard to train. For many years this limited practicla developemnt of arificial neural networks until the adoption of non-linear threshold functons and the development backpropagation.

Defined on page 110

Used on Chap. 6: pages 110, 112, 113, 122, 123; Chap. 7: page 136; Chap. 8: pages 152, 153; Chap. 9: page 192

A single perceptron