Information Theory and Statistical Mechanics II (1957)

Abstract

This sequel had the primary purpose of extending the Maxent formalism to the density matrix, but it also went much more deeply into conceptual matters. Its weakness was in trying to do too much. A single article communicates best if it confines itself to a single topic, but this one touched lightly on a dozen matters, each of which needs a full article to do it justice. In particular, all the ramifications of the ‘Information Game’ and its relation to stochastic theory, could easily fill a book. The article remains, to this day, a source of unfinished fragments of ideas in need of development.