The Setting
Take a measure space , an abelian group written additively, a family of projective characters of on and a dissociated sequence in . Against the characters put the observables , where are measurable maps from to a fixed finite set and is a complex function on with . The question is how large the correlations can be on average over when the process has entropy rate zero, the information carried by its first letters growing sublinearly in .
Projective Characters and Dissociated Sequences
A family of projective characters of an abelian group on a measure space is a family of -measurable complex functions on with almost everywhere for every , with almost everywhere and for every , which is multiplicative up to constant phases: for all there is a complex number of modulus one with (1)✦
A sequence in an abelian group is dissociated if no signed sum of finitely many of its terms with distinct indices vanishes: for every nonempty finite set of indices and every choice of signs , (2)✦
Word Entropy
The Shannon entropy of a function from a finite set to the real numbers is (3)✦ with the natural logarithm and the convention .
The word map of length of a sequence of maps from to a set is the map that sends to the word .✦
The word entropy at length of a sequence of maps from to a finite set is the Shannon entropy of the law of its word map of length under , which gives each word the measure of the set of points sent to it: (4)✦
A sequence of maps from to a finite set has entropy rate zero if as .
The Claim
Theorem 1. Let be a probability space, an abelian group and a family of projective characters of on . Let be a dissociated sequence in , let be measurable maps from to a finite set of entropy rate zero, and let be a complex function on with . Then (5)✦
In the paper, the result reads: if the are projective characters, is dissociated and functions , written in the paper, take values in a fixed finite subset of the closed unit disk with , then .
What Is Checked
The claim (5) is stated over a family of projective characters (1), a dissociated sequence (2) and the word entropy (4), the Shannon entropy (3) of the law of the word map of length .