A conditional independence framework for coherent modularized inference

TL;DR

This paper introduces a formal framework for integrating diverse probabilistic models coherently, ensuring consistent inference across complex, multi-expert systems with evidence updates.

Contribution

It develops a statistical methodology with sufficient conditions to maintain coherence when combining and updating multiple probabilistic models.

Findings

Derived conditions for coherent inference in composite models

Ensured inference consistency before and after evidence incorporation

Provided a formal basis for modular probabilistic modeling

Abstract

Inference in current domains of application are often complex and require us to integrate the expertise of a variety of disparate panels of experts and models coherently. In this paper we develop a formal statistical methodology to guide the networking together of a diverse collection of probabilistic models. In particular, we derive sufficient conditions that ensure inference remains coherent across the composite before and after accommodating relevant evidence.