Measuring multivariate redundant information with pointwise common change in surprisal

TL;DR

This paper introduces a new measure for quantifying multivariate redundant information based on pointwise common change in surprisal, providing a clearer decomposition of information contributions.

Contribution

It proposes a novel redundancy measure using maximum entropy distributions and local co-information, enhancing the Partial Information Decomposition framework.

Findings

The new measure aligns well with existing approaches on example systems.

It effectively decomposes multivariate mutual information into redundant, unique, and synergistic parts.

The approach is applicable to continuous Gaussian variables.

Abstract

The problem of how to properly quantify redundant information is an open question that has been the subject of much recent research. Redundant information refers to information about a target variable S that is common to two or more predictor variables Xi. It can be thought of as quantifying overlapping information content or similarities in the representation of S between the Xi. We present a new measure of redundancy which measures the common change in surprisal shared between variables at the local or pointwise level. We provide a game-theoretic operational definition of unique information, and use this to derive constraints which are used to obtain a maximum entropy distribution. Redundancy is then calculated from this maximum entropy distribution by counting only those local co-information terms which admit an unambiguous interpretation as redundant information. We show how this redundancy measure can be used within the framework of the Partial Information Decomposition (PID) to give an intuitive decomposition of the multivariate mutual information into redundant, unique and synergistic contributions. We compare our new measure to existing approaches over a range of example systems, including continuous Gaussian variables. Matlab code for the measure is provided, including all considered examples.