Part of Advances in Neural Information Processing Systems 33 (NeurIPS 2020)
Karthikeyan Natesan Ramamurthy, Bhanukiran Vinzamuri, Yunfeng Zhang, Amit Dhurandhar
In recent years, post-hoc local instance-level and global dataset-level explainability of black-box models has received a lot of attention. Lesser attention has been given to obtaining insights at intermediate or group levels, which is a need outlined in recent works that study the challenges in realizing the guidelines in the General Data Protection Regulation (GDPR). In this paper, we propose a meta-method that, given a typical local explainability method, can build a multilevel explanation tree. The leaves of this tree correspond to local explanations, the root corresponds to global explanation, and intermediate levels correspond to explanations for groups of data points that it automatically clusters. The method can also leverage side information, where users can specify points for which they may want the explanations to be similar. We argue that such a multilevel structure can also be an effective form of communication, where one could obtain few explanations that characterize the entire dataset by considering an appropriate level in our explanation tree. Explanations for novel test points can be cost-efficiently obtained by associating them with the closest training points. When the local explainability technique is generalized additive (viz. LIME, GAMs), we develop fast approximate algorithm for building the multilevel tree and study its convergence behavior. We show that we produce high fidelity sparse explanations on several public datasets and also validate the effectiveness of the proposed technique based on two human studies -- one with experts and the other with non-expert users -- on real world datasets.