In a series of recent papers, Randall and Foulis report the development of a generalized theory of probability which is based on the concept of a physical operation. A central concept in this theory is that of a generalized sample space. In this paper, we introduce a generalized sample space, which for historial reasons we shall call the Poincaré sphere sample space. We investigate the relationship between this nonclassical sample space and its classical analogs, and find that the key to (...) this relationship is found in the use of idealized (Gedanken) experiments. The Poincaré sphere sample space is seen to describe a “photon” description of polarization, and the classical analogs represent different classical descriptions. We also discuss superposition and entropy for the Poincaré sphere sample space and its classical analogs, and we comment on the application of this generalized theory of probability to problems in the foundations of quantum theory. (shrink)

When the state of a physical system is not fully determined by available data, it should be possible nevertheless to make a systematic guess concerning the unknown state by applying the principles of information theory. The resulting theoretical blend of informational and mechanical constructs should then constitute a modern structure for statistical physics. Such a program has been attempted by a number of authors, most notably Jaynes, with seeming success. However, we demonstrated in a recent publication that the standard list (...) of so-called “mutually exclusive and exhaustive” quantum states that is commonly employed by these authors is in fact not exhaustive. It follows that the information-theoretic foundations of quantum statistics must be reformulated. The present paper discusses the fundamental problems involved and establishes a format for the correct application of information theory to quantum mechanical situations. (shrink)

In previous publications we have criticized the usual application of information theory to quantal situations and proposed a new version of information-theoretic quantum statistics. This paper is the first in a two-part series in which our new approach is applied to the fundamental problem of thermodynamic equilibrium. Part I deals in particular with informational definitions of equilibrium and the identification of thermodynamic analogs in our modified quantum statistics formalism.

The identification of a set of mutually exclusive and exhaustive propositions concerning the states of quantum systems is a corner stone of the information-theoretic foundations of quantum statistics; but the set which is conventionally adopted is in fact incomplete, and is customarily deduced from numerous misconceptions of basic quantum mechanical principles. This paper exposes and corrects these common misstatements. It then identifies a new set of quantum state propositions which is truly exhaustive and mutually exclusive, and which is compatible with (...) the foundations of quantum theory. (shrink)

Many statistical problems, including some of the most important for physical applications, have long been regarded as underdetermined from the standpoint of a strict frequency definition of probability; yet they may appear wellposed or even overdetermined by the principles of maximum entropy and transformation groups. Furthermore, the distributions found by these methods turn out to have a definite frequency correspondence; the distribution obtained by invariance under a transformation group is by far the most likely to be observed experimentally, in the (...) sense that it requires by far the least “skill.” These properties are illustrated by analyzing the famous Bertrand paradox. On the viewpoint advocated here, Bertrand's problem turns out to be well posed after all, and the unique solution has been verified experimentally. We conclude that probability theory has a wider range of useful applications than would be supposed from the standpoint of the usual frequency definitions. (shrink)