Closed-Form Gaussian Estimators for Multi-Source Partial Information Decomposition

TL;DR

This paper introduces the first covariance-based closed-form estimators for multi-source partial information decomposition in continuous Gaussian data, enabling efficient and stable computation of complex information measures.

Contribution

The authors develop novel closed-form Gaussian estimators for multi-source PID, extending beyond two sources and providing key information measures from covariance matrices.

Findings

Estimator is plug-in consistent and affine invariant.

Validated on Gaussian benchmarks with superior efficiency.

Confirmed numerical stability in finite samples.

Abstract

Computing multi-source partial information decomposition (PID) for continuous data is hard: existing closed-form Gaussian estimators are restricted to two source variables, while continuous arbitrary-source estimators are typically learning-based and do not provide closed-form expressions. To address this, we develop closed-form Gaussian estimators for multi-source PID. We provide two-source redundancy, multi-source unique information, the K-th order synergistic effect from source subsets of size K, and the total synergistic effect. The estimators are derived from the conditional-independence-based information measures introduced in our earlier work, under which every quantity reduces to a log-determinant expression in covariance blocks of the system. The resulting estimator is plug-in consistent, affine invariant, source-permutation symmetric, and additive over independent systems. We validate it on a controlled Gaussian benchmark, evaluate its computational efficiency against baselines, and confirm its numerical stability in finite-sample regimes. To our knowledge, this is the first covariance-based closed-form estimator that provides multi-source continuous PID measures.