An overview of the quantitative causality analysis and causal graph reconstruction based on a rigorous formalism of information flow

TL;DR

This paper reviews the development of quantitative causality analysis based on information flow, highlighting its theoretical foundations and applications in artificial intelligence over the past 16 years.

Contribution

It provides a concise overview of the formalism of information flow for causal inference and its reconstruction of causal graphs from data.

Findings

Quantitative causality analysis has been developed independently in physics and AI.

The formalism of information flow enables causal graph reconstruction.

Applications demonstrate the effectiveness of the approach in various AI tasks.

Abstract

Inference of causal relations from data now has become an important field in artificial intelligence. During the past 16 years, causality analysis (in a quantitative sense) has been developed independently in physics from first principles. This short note is a brief summary of this line of work, including part of the theory and several representative applications.