The Original Three Pillars of Early AI Research

The field of Artificial Intelligence was formally founded at the 1956 Dartmouth workshop. Early influential academic institutions included MIT, Stanford, and Carnegie Mellon. Foundational concepts like symbolic AI (exemplified by John McCarthy's Lisp and Advice Taker) and early neural networks (Frank Rosenblatt's Perceptron) emerged during this pioneering era.

The top 3

  1. Symbolic AI: Logic, Rules, and Early Systems: This foundational approach, also known as GOFAI, focused on explicit knowledge representation and logical reasoning, exemplified by the 1956 Logic Theorist and the General Problem Solver programs by Newell and Simon.
  2. Connectionism: Neural Nets, Pattern Learning: Inspired by the human brain, this paradigm emphasizes learning from data through artificial neural networks, with early contributions like Frank Rosenblatt's Perceptron (1957) and Oliver Selfridge's Pandemonium model (1959).
  3. Cybernetics: Control, Communication, Feedback: A crucial precursor to AI, cybernetics, pioneered by Norbert Wiener in 1948, explored control and communication in biological, mechanical, and electronic systems, providing theoretical foundations for self-regulating machines.

Sources

Open the full topic