Philosophy of AI
The philosophy of artificial intelligence considers the question „Chan machines think?”. The first man who tried to answer this question was Alan Turing in his classic paper from 1950 „Computing machinery and inteligence”. Since then several answers have been given.
The Dartmouth proposal: „Every aspect of learning or any other feature of intelligence can be so precisley described that a machine can be made to simulate it” .
Hubert Dreyfus argued that, on the countrary, human experties depends on unconcious instinct rather that councious symbol manipulation.
Traditional Symbloic AI
When acces to digital computers became posible AI researchers began to explore the possibility that a human intelligente could be reduced to simblo manipulation.
The research was centured two three institutions Stanford, CMU and MIT. Each one of these three institutions developed their own style of research.
Cognitive simulation
Economist Herbert Simon and Alan Newell studied human skils in problem solving and atempted to formalize them. Their work leid the fundations of the field of artificial intelligence and cognitive science, operations research and management science. Their research team studyed the similarities between human problem resolving skils and the program that they were developing.
Logical AI
Unlike Simon and Alan, Jhon McCharty belived that machines did not need to simulate human thought, but should instead find the essence of abstract reasoning and problem solving.
His laboratory an Stratford focussed on using logic to solve wide variety of problems , including knowledge representation, planing and learning. Work in logic led to the developing of the programming language Prolog and the science of logic programming.
„Scruffy” simbloci AI
Researchers from MIT found that solving difficult problems in vission and natural languege processing would require ad-hoc solutions. They argued that there is no easy answer , no simple and general principle that would capture all the aspects of intelligent behavior.
Knowledge based AI
When computers with large memories became available in 1970 AI researchers to build knowledge into Ai applications. This knowledge development led to the development expert system the most succesful form of AI.
The knowledge revolution was also driven by the realization that truli enormous of amounts knowledge would be required by many simple AI applications.