With the ever-increasing popularity and applications of graph neural networks, several proposals have been made to interpret and understand the decisions of a GNN model. Explanations for a GNN model differ in principle from other input settings. It is important to attribute the decision to input features and other related instances connected by the graph structure. We find that the previous explanation generation approaches that maximize the mutual information between the label distribution produced by the GNN model and the explanation to be restrictive. Specifically, existing approaches do not enforce explanations to be predictive, sparse, or robust to input perturbations.In this paper, we lay down some of the fundamental principles that an explanation method for GNNs should follow and introduce a metric fidelity as a measure of the explanation's effectiveness. We propose a novel approach Zorro based on the principles from rate-distortion theory that uses a simple combinatorial procedure to optimize for fidelity. Extensive experiments on real and synthetic datasets reveal that Zorro produces sparser, stable, and more faithful explanations than existing GNN explanation approaches.
This study confirmed that replacing mOPV1 with bOPV in campaigns was successful in maintaining very high population immunity to type 1 poliovirus and substantially decreasing the immunity gap to type 3 poliovirus.
Since 2004, efforts to improve poliovirus detection have significantly increased the volume of specimen testing from acute flaccid paralysis (AFP) patients in India. One option to decrease collection and testing burden would be collecting only a single stool specimen instead of two. We investigated stool specimen sensitivity for poliovirus detection in India to estimate the contribution of the second specimen. We reviewed poliovirus isolation data for 303984 children aged <15 years with AFP during 2000-2010. Using maximum-likelihood estimation, we determined specimen sensitivity of each stool specimen, combined sensitivity of both specimens, and sensitivity added by the second specimen. Of 5184 AFP patients with poliovirus isolates, 382 (7.4%) were identified only by the second specimen. Sensitivity was 91.4% for the first specimen and 84.5% for the second specimen; the second specimen added 7.3% sensitivity, giving a combined sensitivity of 98.7%. Combined sensitivity declined, and added sensitivity increased, as the time from paralysis onset to stool collection increased (P = 0.032). The sensitivity added by the second specimen is important to detect the last chains of poliovirus transmission and to achieve certification of polio eradication. For sensitive surveillance, two stool specimens should continue to be collected from each AFP patient in India.
Privacy and interpretability are two of the important ingredients for achieving trustworthy machine learning. We study the interplay of these two aspects in graph machine learning through graph reconstruction attacks. The goal of the adversary here is to reconstruct the graph structure of the training data given access to model explanations. Based on the different kinds of auxiliary information available to the adversary, we propose several graph reconstruction attacks. We show that additional knowledge of post-hoc feature explanations substantially increases the success rate of these attacks. Further, we investigate in detail the differences between attack performance with respect to three different classes of explanation methods for graph neural networks: gradient-based, perturbationbased, and surrogate model-based methods. While gradient-based explanations reveal the most in terms of the graph structure, we find that these explanations do not always score high in utility. For the other two classes of explanations, privacy leakage increases with an increase in explanation utility. Finally, we propose a defense based on a randomized response mechanism for releasing the explanations which substantially reduces the attack success rate. Our anonymized code is available at xxxxxxxxx.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.