The powerful new technique of maximum entropy, which has its roots in probability theory, provides a much needed extension of the established principles of rational inference in the sciences. It allows the interpretation of incomplete and noisy data, to provide a description of the underlying physical system, and has found application in both practical and theoretical studies ranging form image enhancement to nuclear physics, from statistical mechanics to economics.This book is a diverse collection of introductory articles based on a series of interdisciplinary lectures covering the fundamentals of the maximum entropy approach and Bayesian methods, as well as the application of the method to various problems of data analysis in the physical sciences.