Timezone: »

 
Poster
On Completeness-aware Concept-Based Explanations in Deep Neural Networks
Chih-Kuan Yeh · Been Kim · Sercan Arik · Chun-Liang Li · Tomas Pfister · Pradeep Ravikumar

Tue Dec 08 09:00 PM -- 11:00 PM (PST) @ Poster Session 2 #640
in Poster Session 3 »

Human explanations of high-level decisions are often expressed in terms of key concepts the decisions are based on. In this paper, we study such concept-based explainability for Deep Neural Networks (DNNs). First, we define the notion of \emph{completeness}, which quantifies how sufficient a particular set of concepts is in explaining a model's prediction behavior based on the assumption that complete concept scores are sufficient statistics of the model prediction. Next, we propose a concept discovery method that aims to infer a complete set of concepts that are additionally encouraged to be interpretable, which addresses the limitations of existing methods on concept explanations. To define an importance score for each discovered concept, we adapt game-theoretic notions to aggregate over sets and propose \emph{ConceptSHAP}. Via proposed metrics and user studies, on a synthetic dataset with apriori-known concept explanations, as well as on real-world image and language datasets, we validate the effectiveness of our method in finding concepts that are both complete in explaining the decisions and interpretable.

Keywords: Gaussian Processes · Distributed Inference · Probabilistic Methods · Probabilistic Methods -> Bayesian Nonparametrics; Probabilistic Methods

Author Information

Chih-Kuan Yeh (Carnegie Mellon University)
Been Kim (Google)
Sercan Arik (Google)
Chun-Liang Li (Google)
Tomas Pfister (Google)
Pradeep Ravikumar (Carnegie Mellon University)

More from the Same Authors