A New Perspective on Learning Context-Specific Independence

TL;DR

This paper introduces a novel method for learning context-specific independencies in probabilistic graphical models by combining neural network-based CPT representations with quantization into arithmetic circuits, enabling efficient inference and leveraging machine learning and explainable AI tools.

Contribution

It proposes a new approach that first learns a neural network representation of CPTs and then quantizes it into an arithmetic circuit, contrasting with traditional variable-splitting methods.

Findings

Effective learning of CSIs from data using neural networks.

Quantization into arithmetic circuits enables efficient inference.

Empirical results show advantages over traditional methods.

Abstract

Local structure such as context-specific independence (CSI) has received much attention in the probabilistic graphical model (PGM) literature, as it facilitates the modeling of large complex systems, as well as for reasoning with them. In this paper, we provide a new perspective on how to learn CSIs from data. We propose to first learn a functional and parameterized representation of a conditional probability table (CPT), such as a neural network. Next, we quantize this continuous function, into an arithmetic circuit representation that facilitates efficient inference. In the first step, we can leverage the many powerful tools that have been developed in the machine learning literature. In the second step, we exploit more recently-developed analytic tools from explainable AI, for the purposes of learning CSIs. Finally, we contrast our approach, empirically and conceptually, with more traditional variable-splitting approaches, that search for CSIs more explicitly.