? Review of ‘ Reliable Reasoning ’ by Gilbert Harman and Sanjeev Kulkarni

The aim of the book is to give a non-technical introduction to statistical learning theory at undergraduate level. Statistical learning theory is concerned with the reliability of rules for classifying a new case—e.g., diagnosing a disease in a new patient—on the basis of other features of the case and a large stock of past cases and their features and classifications. The book is based on a course on learning theory and epistemology given to undergraduate students in electrical engineering and in philosophy at Princeton. It is a short book, with four chapters. The first chapter is on 'the problem of induction'. Note though, that the book is not concerned with what philosophers normally take to be 'the problem of induction'—the problem of justifying induction—but rather with the problem of assessing the reliability of inductive rules. The chapter then sets out to debunk the commonly held view that there are two sorts of reasoning, deductive and inductive, and two sorts of arguments, deduc-tive and inductive, by means of arguments to which I will return below. Chapter 2 introduces the idea of using enumerative induction to learn classification rules, and for estimating the values of continuous variables. It introduces the VC-dimension of a set of classification rules, which is perhaps the most important concept in statistical learning theory. Chapter 3 discusses induction rules which work by trying to rank hypotheses by their simplicity. Chapter 4 discusses applications of statistical learning theory to neural networks and support vector machines in machine learning. The book does a good job at presenting the main ideas underlying statistical learning theory. Members at our interdisciplinary Centre for Reasoning at the University of Kent read the book in a reading group and found that those who were unfamiliar with statistical learning theory were able to grasp the central intuitions, while those who were familiar with statistical learning theory had some controversial philosophical claims to get their teeth into. Philosophical topics include the nature of induction, Goodman's new problem of induction, simplicity, mental processes for reasoning, and moral particularism. Note that the book is intentionally light on detail, so it may help to have someone familiar with both the statistical and the philosophical details on hand to fill in gaps and answer students' questions. This restricts the applicability of the book somewhat. While the book largely succeeds in its aim of presenting the rudiments of statistical learning theory in …