A Functional Information Perspective on Model Interpretation

Itai Gat · Nitay Calderon · Roi Reichart · Tamir Hazan

Room 309
[ Abstract ] [ Livestream: Visit Deep Learning/MISC ]
Thu 21 Jul 10:35 a.m. — 10:40 a.m. PDT
[ Slides [ Paper PDF

Contemporary predictive models are hard to interpret as their deep nets exploit numerous complex relations between input elements. This work suggests a theoretical framework for model interpretability by measuring the contribution of relevant features to the functional entropy of the network with respect to the input. We rely on the log-Sobolev inequality that bounds the functional entropy by the functional Fisher information with respect to the covariance of the data. This provides a principled way to measure the amount of information contribution of a subset of features to the decision function. Through extensive experiments, we show that our method surpasses existing interpretability sampling-based methods on various data signals such as image, text, and audio.

Chat is not available.