Reliable Graph Neural Networks via Robust Aggregation

10/29/2020
by   Simon Geisler, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

08/20/2019

Robust Graph Neural Network Against Poisoning Attacks via Transfer Learning

Graph neural networks (GNNs) are widely used in many applications. Howev...
06/28/2019

Certifiable Robustness and Robust Training for Graph Convolutional Networks

Recent works show that Graph Neural Networks (GNNs) are highly non-robus...
06/06/2021

On Local Aggregation in Heterophilic Graphs

Many recent works have studied the performance of Graph Neural Networks ...
10/31/2019

Certifiable Robustness to Graph Perturbations

Despite the exploding interest in graph neural networks there has been l...
10/12/2021

Scalable Consistency Training for Graph Neural Networks via Self-Ensemble Self-Distillation

Consistency training is a popular method to improve deep learning models...
12/14/2021

Robust Graph Neural Networks via Probabilistic Lipschitz Constraints

Graph neural networks (GNNs) have recently been demonstrated to perform ...
05/31/2019

Discriminative structural graph classification

This paper focuses on the discrimination capacity of aggregation functio...