Textbook of AI

1894–1964, Mathematician, founder of cybernetics

Also known as: N. Wiener

Norbert Wiener was an American child prodigy and mathematician, he received his Harvard PhD in mathematical logic at 18, who founded the field of cybernetics and gave it its name. His 1948 book Cybernetics: or Control and Communication in the Animal and the Machine synthesised work from his wartime project on automatic anti-aircraft fire control, his collaborations with neurophysiologists, and his earlier mathematical work on stochastic processes and harmonic analysis. The central idea was that feedback, using a system's output to regulate its input, provides a unified framework for understanding goal-directed behaviour in machines, organisms, and societies.

Cybernetics directly anticipated reinforcement learning, control theory, robotics, and much of what we now call AI. The Macy Conferences on Cybernetics (1946–1953), which Wiener helped lead with McCulloch, drew together McCulloch, Pitts, von Neumann, Shannon, Margaret Mead, Gregory Bateson and others; they were the first sustained interdisciplinary forum for the questions that would become AI, computer science, cognitive science and information theory.

Wiener's mathematical legacy is independent of and arguably even larger than cybernetics. The Wiener process (Brownian motion) underpins stochastic calculus and modern mathematical finance; the Wiener filter is the optimal linear filter for stationary signals in noise. He spent the bulk of his career at MIT and was a passionate advocate against the militarisation of science, publicly refusing post-war military contracts and writing extensively on the social risks of automation in The Human Use of Human Beings (1950).