"Why Should I Trust You?": Explaining the Predictions of any Classifier

Despite widespread adoption, machine learning algorithms mostly remain black-boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one. In this talk, I will describe the LIME algorithm, an explanation technique that explains the predictions of ANY classifier in an interpretable and faithful manner, by learning an interpretable model locally around the prediction. We also propose a method to explain whole models by presenting representative explanations in a non-redundant way, framing the task as a submodular optimization problem. We demonstrate the flexibility of these methods by explaining different models for text (e.g. random forests) and image classification (e.g. neural networks). We show the utility of explanations via experiments with human subjects, on various scenarios that require trust: deciding if one should trust a prediction, choosing between two algorithms, improving an untrustworthy classifier, and characterizing exactly why a classifier should not be trusted. Work with Marco Tulio Ribeiro (University of Washington) and Carlos Guestrin (Apple/University of Washington)