Neural Bayesian Network Understudy

TL;DR

This paper explores combining neural networks with Bayesian Network principles to incorporate causal knowledge, enabling neural models to approximate Bayesian Network functionalities and causal structures.

Contribution

It introduces methods to train neural networks that output conditional probabilities and encode causal independence relations, acting as understudies to Bayesian Networks.

Findings

Neural networks can approximate Bayesian Network probabilistic outputs.

Proposed training strategies encode causal independence relations.

Initial results show neural models mimic Bayesian Network properties.

Abstract

Bayesian Networks may be appealing for clinical decision-making due to their inclusion of causal knowledge, but their practical adoption remains limited as a result of their inability to deal with unstructured data. While neural networks do not have this limitation, they are not interpretable and are inherently unable to deal with causal structure in the input space. Our goal is to build neural networks that combine the advantages of both approaches. Motivated by the perspective to inject causal knowledge while training such neural networks, this work presents initial steps in that direction. We demonstrate how a neural network can be trained to output conditional probabilities, providing approximately the same functionality as a Bayesian Network. Additionally, we propose two training strategies that allow encoding the independence relations inferred from a given causal structure into the neural network. We present initial results in a proof-of-concept setting, showing that the neural model acts as an understudy to its Bayesian Network counterpart, approximating its probabilistic and causal properties.