Learning Credible Models

TL;DR

This paper introduces a new regularization method called EYE that enhances the credibility of interpretable linear models by incorporating expert knowledge, leading to more trustworthy and clinically relevant predictions.

Contribution

The paper proposes the EYE regularization penalty that integrates expert knowledge into linear models to improve their credibility without sacrificing accuracy.

Findings

EYE models are more credible than other regularization methods.

EYE achieves better alignment with known domain knowledge.

EYE maintains strong predictive performance in real-world tasks.

Abstract

In many settings, it is important that a model be capable of providing reasons for its predictions (i.e., the model must be interpretable). However, the model's reasoning may not conform with well-established knowledge. In such cases, while interpretable, the model lacks \textit{credibility}. In this work, we formally define credibility in the linear setting and focus on techniques for learning models that are both accurate and credible. In particular, we propose a regularization penalty, expert yielded estimates (EYE), that incorporates expert knowledge about well-known relationships among covariates and the outcome of interest. We give both theoretical and empirical results comparing our proposed method to several other regularization techniques. Across a range of settings, experiments on both synthetic and real data show that models learned using the EYE penalty are significantly more credible than those learned using other penalties. Applied to a large-scale patient risk stratification task, our proposed technique results in a model whose top features overlap significantly with known clinical risk factors, while still achieving good predictive performance.