In this paper, we propose ContrastNet, an effective adversarial defense framework for GNN. In particular, we propose an adversarial contrastive learning method.

Abstract—Graph Neural Network (GNN), as a powerful repre- sentation learning model on graph data, attracts much attention across various disciplines.

In this paper, we propose ContrastNet, an effective adversarial defense framework for GNN. In particular, we propose an adversarial contrastive learning method.

Feb 25, 2021 · In the contrastive learning framework, we introduce a new method that increases the adversarial robustness of the learned representations ...

Duration: 18:46
Posted: Jun 8, 2023

GROC builds on top of previous work in contrastive learning on graphs, aiming to improve graph neural networks' robustness against adversarial attacks. 3.1 ...

This work proposes a simple but effective method, Adversarial Graph Contrastive Learning (ArieL), to extract informative contrastive samples within a ...

Jan 31, 2022 · We propose a novel Graph Adversarial Contrastive Learning framework (GraphACL) by introducing adversarial augmentations into graph self-supervised learning.

This repository contains Attack-related papers, Defense-related papers, Robustness Certification papers, etc., ranging from 2017 to 2021.

This paper proposes a method called GraphDefense to defend against the adversarial perturbations of graph convolutional networks, and shows that with ...