Machine Learning from Examples: A Non-Inductivist Analysis

It has been suggested that AI investigations of mechanical learning undermine sweeping anti-inductivist views in the theory of knowledge and the philosophy of science. In particular, it is claimed that some mechanical learning systems perform epistemically justified inductive generalization and prediction. Contrary to this view, it is argued that no trace of such epistemic justification is to be found within a rather representative class of learning agents drawn from machine learning and robotics. Moreover, an alternative deductive account of these learning procedures is outlined. Finally, the opportunity of developing an induction-free logical analysis of non-monotonic reasoning in autonomous learning agents – capable of advancing and revising learning or background hypotheses – is emphasized by a broad reflection on some families of non-monotonic, albeit deductive, consequence relations.1 We are grateful to Jeff Paris and David Makinson for useful comments and suggestions.