Premium
On Cheeseman
Author(s) -
Israel David
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.tb00100.x
Subject(s) - center (category theory) , citation , library science , computer science , artificial intelligence , information retrieval , crystallography , chemistry
Cheeseman wants to convince us that Bayesianism is the proper framewok for research in representation and reasoning in AI, or slightly more cautiously, that it is an ineliminable and major part of any such framework. I am not convinced. My reasons, though, are most certainly not that I think that reasoning, commonsense or otherwise, is typically logical or deductively sound. Moreover, I share his doubts (or at any rate, I have doubts of my own) about the various proposed nonmonotonic extensions, for example, to classical first-order logic. But here I want neither to praise “logic” nor to bury it; nor even to explain my reservations about Bayesianism. I want, mther to correct what seem to me misconceptions about both logic and subjective probability, misconceptions that are central to Cheeseman’s case.’