Poster
Streaming Weak Submodularity: Interpreting Neural Networks on the Fly
Ethan Elenberg · Alexandros Dimakis · Moran Feldman · Amin Karbasi

Tue Dec 5th 06:30 -- 10:30 PM @ Pacific Ballroom #155 #None

In many machine learning applications, it is important to explain the predictions of a black-box classifier. For example, why does a deep neural network assign an image to a particular class? We cast interpretability of black-box classifiers as a combinatorial maximization problem and propose an efficient streaming algorithm to solve it subject to cardinality constraints. By extending ideas from Badanidiyuru et al. [2014], we provide a constant factor approximation guarantee for our algorithm in the case of random stream order and a weakly submodular objective function. This is the first such theoretical guarantee for this general class of functions, and we also show that no such algorithm exists for a worst case stream order. Our algorithm obtains similar explanations of Inception V3 predictions 10 times faster than the state-of-the-art LIME framework of Ribeiro et al. [2016].

Author Information

Ethan Elenberg (ASAPP)
Alex Dimakis (University of Texas, Austin)
Moran Feldman (Open University of Israel)
Amin Karbasi (Yale)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors