Premium
Probability, truth, and logic: reply to Cheeseman
Author(s) -
Bundy Alan
Publication year - 1988
Publication title -
computational intelligence
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.353
H-Index - 52
eISSN - 1467-8640
pISSN - 0824-7935
DOI - 10.1111/j.1467-8640.1988.tb00093.x
Subject(s) - bridge (graph theory) , citation , artificial intelligence , computer science , natural language processing , information retrieval , operations research , library science , medicine , engineering
tive reasoning betrays the same attitude that Hilbert’s 6th problem does. (This problem, posed before Einstein’s work became public, encouraged mathematicians to axiomatize physics.) Both are rooted in the belief that science is merely an extension of mathematics, philosophy, and logic. Herbert Simon, on the other hand, studies the psychology of humans doing science, and not how they ought to do it. He finds this a fruitful source of ideas for building computer models of the scientific reasoning process. In short I believe we must build theories that describe how humans do induction-these will be a better source of inspiration than philosophical argumentation about normative theories. Acknowledgements This work has been supported by the Natural Sciences and Engineering Research Council of Canada. Denis GagnC, Eric Neufeld, and Maarten van Emden suggested welcome improvements.