Are Logistic Models Really Interpretable?

TL;DR

This paper introduces Linearised Additive Models (LAMs) as a more interpretable alternative to logistic regression, demonstrating their effectiveness through user studies and comparable predictive performance on financial data.

Contribution

The paper proposes LAMs as a novel, retraining-free approximation of logistic models that enhances interpretability without sacrificing accuracy.

Findings

Skilled users struggle to interpret LR models accurately.

LAMs improve interpretability in model reasoning tasks.

LAMs maintain similar ROC-AUC and calibration as LR models.

Abstract

The demand for open and trustworthy AI models points towards widespread publishing of model weights. Consumers of these model weights must be able to act accordingly with the information provided. That said, one of the simplest AI classification models, Logistic Regression (LR), has an unwieldy interpretation of its model weights, with greater difficulties when extending LR to generalised additive models. In this work, we show via a User Study that skilled participants are unable to reliably reproduce the action of small LR models given the trained parameters. As an antidote to this, we define Linearised Additive Models (LAMs), an optimal piecewise linear approximation that augments any trained additive model equipped with a sigmoid link function, requiring no retraining. We argue that LAMs are more interpretable than logistic models -- survey participants are shown to solve model reasoning tasks with LAMs much more accurately than with LR given the same information. Furthermore, we show that LAMs do not suffer from large performance penalties in terms of ROC-AUC and calibration with respect to their logistic counterparts on a broad suite of public financial modelling data.