Reliable Graph Neural Networks via Robust Aggregation

This page links to additional material for our paper

Reliable Graph Neural Networks via Robust Aggregation
by Simon Geisler, Daniel Zügner and Stephan Günnemann
Published at the Neural Information Processing Systems (NeurIPS) 2020


[Paper | GitHub | Video (Slideslive)Google Colab Notebook]


Perturbations targeting the graph structure have proven to be extremely effective in reducing the performance of Graph Neural Networks (GNNs), and traditional defenses such as adversarial training do not seem to be able to improve robustness. This work is motivated by the observation that adversarially injected edges effectively can be viewed as additional samples to a node’s neighborhood aggregation function, which results in distorted aggregations accumulating over the layers. Conventional GNN aggregation functions, such as a sum or mean, can be distorted arbitrarily by a single outlier. We propose a robust aggregation function motivated by the field of robust statistics. Our approach exhibits the largest possible breakdown point of 0.5, which means that the bias of the aggregation is bounded as long as the fraction of adversarial edges of a node is less than 50%. Our novel aggregation function, Soft Medoid, is a fully differentiable generalization of the Medoid and therefore lends itself well for end-to-end deep learning. Equipping a GNN with our aggregation improves the robustness with respect to structure perturbations on Cora ML by a factor of 3 (and 5.5 on Citeseer) and by a factor of 8 for low-degree nodes.


Please cite our paper if you use the method in your own work:

title = {Reliable Graph Neural Networks via Robust Aggregation},
author = {Geisler, Simon and Z{\"{u}}gner, Daniel and G{\"{u}}nnemann, Stephan},
booktitle={Neural Information Processing Systems, {NeurIPS}},

year = {2020}