Bayesian Stress Testing of Models in a Classification Hierarchy

TL;DR

This paper introduces a Bayesian framework for stress testing hierarchical machine learning models, enhancing confidence in their performance before deployment across various applications.

Contribution

It presents a novel Bayesian approach to model interactions in hierarchical ML systems, enabling effective stress testing and performance assessment.

Findings

Framework successfully applied to toy and real datasets

Improves understanding of model interactions in hierarchies

Facilitates pre-deployment confidence in ML solutions

Abstract

Building a machine learning solution in real-life applications often involves the decomposition of the problem into multiple models of various complexity. This has advantages in terms of overall performance, better interpretability of the outcomes, and easier model maintenance. In this work we propose a Bayesian framework to model the interaction amongst models in such a hierarchy. We show that the framework can facilitate stress testing of the overall solution, giving more confidence in its expected performance prior to active deployment. Finally, we test the proposed framework on a toy problem and financial fraud detection dataset to demonstrate how it can be applied for any machine learning based solution, regardless of the underlying modelling required.