A Framework for Auditing Multilevel Models using Explainability Methods

TL;DR

This paper proposes an audit framework for multilevel models focusing on model performance, fairness, and explainability, utilizing explainability methods like SHAP and LIME, and evaluates their effectiveness with real data.

Contribution

It introduces a comprehensive audit framework for regression multilevel models, incorporating KPIs and risk assessment, and compares explainability methods for transparency evaluation.

Findings

SHAP and LIME underperform in accuracy for feature importance

The framework assesses fairness and transparency comprehensively

Open source dataset demonstrates the framework's practical application

Abstract

Applications of multilevel models usually result in binary classification within groups or hierarchies based on a set of input features. For transparent and ethical applications of such models, sound audit frameworks need to be developed. In this paper, an audit framework for technical assessment of regression MLMs is proposed. The focus is on three aspects, model, discrimination, and transparency and explainability. These aspects are subsequently divided into sub aspects. Contributors, such as inter MLM group fairness, feature contribution order, and aggregated feature contribution, are identified for each of these sub aspects. To measure the performance of the contributors, the framework proposes a shortlist of KPIs. A traffic light risk assessment method is furthermore coupled to these KPIs. For assessing transparency and explainability, different explainability methods (SHAP and LIME) are used, which are compared with a model intrinsic method using quantitative methods and machine learning modelling. Using an open source dataset, a model is trained and tested and the KPIs are computed. It is demonstrated that popular explainability methods, such as SHAP and LIME, underperform in accuracy when interpreting these models. They fail to predict the order of feature importance, the magnitudes, and occasionally even the nature of the feature contribution. For other contributors, such as group fairness and their associated KPIs, similar analysis and calculations have been performed with the aim of adding profundity to the proposed audit framework. The framework is expected to assist regulatory bodies in performing conformity assessments of AI systems using multilevel binomial classification models at businesses. It will also benefit businesses deploying MLMs to be future proof and aligned with the European Commission proposed Regulation on Artificial Intelligence.