Model-Agnostic Interpretability of Machine Learning

TL;DR

This paper advocates for model-agnostic interpretability methods that treat machine learning models as black boxes, offering flexible explanations applicable across various models and enhancing understanding, debugging, and user trust.

Contribution

It reviews the importance of model-agnostic explanations in machine learning and discusses the LIME approach as a recent solution to interpretability challenges.

Findings

Model-agnostic explanations improve understanding of black-box models.

LIME provides local, interpretable explanations for any classifier.

Such methods enhance debugging and user trust in machine learning systems.

Abstract

Understanding why machine learning models behave the way they do empowers both system designers and end-users in many ways: in model selection, feature engineering, in order to trust and act upon the predictions, and in more intuitive user interfaces. Thus, interpretability has become a vital concern in machine learning, and work in the area of interpretable models has found renewed interest. In some applications, such models are as accurate as non-interpretable ones, and thus are preferred for their transparency. Even when they are not accurate, they may still be preferred when interpretability is of paramount importance. However, restricting machine learning to interpretable models is often a severe limitation. In this paper we argue for explaining machine learning predictions using model-agnostic approaches. By treating the machine learning models as black-box functions, these approaches provide crucial flexibility in the choice of models, explanations, and representations, improving debugging, comparison, and interfaces for a variety of users and models. We also outline the main challenges for such methods, and review a recently-introduced model-agnostic explanation approach (LIME) that addresses these challenges.