No label defined (Q4817201)

The author correctly makes a sharp distinction between an approach to inductive logic that takes it to be part of {logic}, and an approach that makes use of a subjectivist notion of probability in an attempt to make sense of the way in which people reason inductively. The former is logic; the latter is psychology. The author adds a third study: the recent study in computational learning theory of ways of converging to the truth. In the first category Carnap (and only Carnap) is treated extensively. The second section states de Finetti's representation theorem and briefly discusses ``grue''. The third section does no more than briefly mention some results in computational learning theory. The first section is clear and informative to the nonprofessional; the other two sections are shorter, sketchier, and more difficult to understand. For the entire collection see [Zbl 1031.03004].