
Robust Graph Neural Network Against Poisoning Attacks via Transfer Learning
Graph neural networks (GNNs) are widely used in many applications. Howev...
read it

Certifiable Robustness and Robust Training for Graph Convolutional Networks
Recent works show that Graph Neural Networks (GNNs) are highly nonrobus...
read it

On Local Aggregation in Heterophilic Graphs
Many recent works have studied the performance of Graph Neural Networks ...
read it

Certifiable Robustness to Graph Perturbations
Despite the exploding interest in graph neural networks there has been l...
read it

RedundancyFree Computation Graphs for Graph Neural Networks
Graph Neural Networks (GNNs) are based on repeated aggregations of infor...
read it

Optimal Transport Graph Neural Networks
Current graph neural network (GNN) architectures naively average or sum ...
read it

Discriminative structural graph classification
This paper focuses on the discrimination capacity of aggregation functio...
read it
Reliable Graph Neural Networks via Robust Aggregation
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 endtoend 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 lowdegree nodes.
READ FULL TEXT
Comments
There are no comments yet.