Artificial intelligence · 1958

Perceptron

Rosenblatt's machine learns to classify images by adjusting weights from its own mistakes.

Frank Rosenblatt described the perceptron in 1958 and built the Mark I Perceptron at Cornell, with a 20x20 photocell retina and motor-driven potentiometers as weights. Its learning rule is guaranteed to converge on linearly separable data. It is the direct ancestor of modern neural networks.

See it in the living tree Wikipedia

Arcs above join it to what it stood on, arcs below to what it led to. The bar places this stretch on the tree's whole timeline.

What it built on 2

Hebbian learning 1949
McCulloch-Pitts neuron 1943

What it led to 5

ADALINE 1960
Perceptrons (book) 1969
Neocognitron 1980
Backpropagation 1986
Support vector machine 1995

Related measurements

The dashed line marks when it first worked.

Training compute of notable AI models Source: Epoch AI, Notable AI Models (CC BY 4.0)

Computation per constant dollar, 1850-2006 Latest: 7.3B million units of computer power per 2006 $ (2003) Source: Nordhaus 2007, Two Centuries of Productivity Growth in Computing, J. Econ. Hist. 67(1), Table 6