Historical background: First generation neural networks

Bomb

Toy

• Perceptrons (~1960) used a layer of hand- coded features and tried to recognize objects by learning how to weight these features. – There was a neat learning algorithm for adjusting the weights. – But perceptrons are fundamentally limited in what they can learn to do.

output units e.g. class labels

non-adaptive hand-coded features

input units e.g. pixels

Sketch of a typical perceptron from the 1960’s