Algorithmic information theory

Algorithmic information theory (AIT) is a branch of theoretical computer science that concerns itself with the relationship between computation and information of computably generated objects (as opposed to stochastically generated), such as strings or any other data structure. In other words, it is shown within algorithmic information theory that computational incompressibility "mimics" (except for a constant that only depends on the chosen universal programming language) the relations or inequalities found in information theory.^[1] According to Gregory Chaitin, it is "the result of putting Shannon's information theory and Turing's computability theory into a cocktail shaker and shaking vigorously."^[2]

Besides the formalization of a universal measure for irreducible information content of computably generated objects, some main achievements of AIT were to show that: in fact algorithmic complexity follows (in the self-delimited case) the same inequalities (except for a constant^[3]) that entropy does, as in classical information theory;^[1] randomness is incompressibility;^[4] and, within the realm of randomly generated software, the probability of occurrence of any data structure is of the order of the shortest program that generates it when running on a universal machine.^[5]

AIT principally studies measures of irreducible information content of strings (or other data structures). Because most mathematical objects can be described in terms of strings, or as the limit of a sequence of strings, it can be used to study a wide variety of mathematical objects, including integers. One of the main motivations behind AIT is the very study of the information carried by mathematical objects as in the field of metamathematics, e.g., as shown by the incompleteness results mentioned below. Other main motivations came from surpassing the limitations of classical information theory for single and fixed objects, formalizing the concept of randomness, and finding a meaningful probabilistic inference without prior knowledge of the probability distribution (e.g., whether it is independent and identically distributed, Markovian, or even stationary). In this way, AIT is known to be basically founded upon three main mathematical concepts and the relations between them: algorithmic complexity, algorithmic randomness, and algorithmic probability.^[6][4]