Link prediction in dynamic graphs (LPDG) is an important research problem that has diverse applications such as online recommendations, studies on disease contagion, organizational studies, etc. Various LPDG methods based on graph embedding and graph neural networks have been recently proposed and achieved state-of-the-art performance. In this paper, we study the vulnerability of LPDG methods and propose the first practical black-box evasion attack. Specifically, given a trained LPDG model, our attack aims to perturb the graph structure, without knowing to model parameters, model architecture, etc., such that the LPDG model makes as many wrong predicted links as possible. We design our attack based on a stochastic policy-based RL algorithm. Moreover, we evaluate our attack on three real-world graph datasets from different application domains. Experimental results show that our attack is both effective and efficient.
Graph neural networks (GNNs) have achieved state-of-the-art performance in many graph-related tasks, e.g., node classification. However, recent works show that GNNs are vulnerable to evasion attacks, i.e., an attacker can slightly perturb the graph structure to fool GNN models. Existing evasion attacks to GNNs have several key drawbacks: 1) they are limited to attack two-layer GNNs; 2) they are not efficient; or/and 3) they need to know GNN model parameters. We address the above drawbacks in this paper and propose an influence-based evasion attack against GNNs. Specifically, we first introduce two influence functions, i.e., feature-label influence and label influence, that are defined on GNNs and label propagation (LP), respectively. Then, we build a strong connection between GNNs and LP in terms of influence. Next, we reformulate the evasion attack against GNNs to be related to calculating label influence on LP, which is applicable to multi-layer GNNs and does not need to know the GNN model. We also propose an efficient algorithm to calculate label influence. Finally, we evaluate our influence-based attack on three benchmark graph datasets. Our experimental results show that, compared to state-of-the-art attack, our attack can achieve comparable attack performance, but has a 5-50x speedup when attacking two-layer GNNs. Moreover, our attack is effective to attack multi-layer GNNs.
Graph neural networks (GNNs) have achieved state-of-the-art performance in many graph-related tasks, e.g., node classification. However, recent works show that GNNs are vulnerable to evasion attacks, i.e., an attacker can perturb the graph structure to fool trained GNN models. Existing evasion attacks to GNNs have two key drawbacks. First, perturbing the graph structure to fool GNN models is essentially a binary optimization problem, while it is often solved via approximate algorithms with sub-optimal solutions. Second, existing attacks are only applicable to two-layer GNNs. In this paper, we aim to address the above drawbacks and propose to attack GNNs via influence function, a completely different perspective from existing works. Specifically, we first build the connection between GNNs and label propagation in terms of influence function. Then, instead of solving an approximate algorithm, we reformulate the attack to be related to (label) influence, which is applicable to multi-layer GNNs and whose solution can be calculated directly. We evaluate our attack on various benchmark graph datasets. Experimental results demonstrate that, compared to state-of-the-art attack, our attack can achieve higher attack success rate and has a 10-100x speedup when attacking two-layer GNNs. Moreover, our attack is also very effective to attack multi-layer GNNs.
Federated learning is a popular distributed machine learning paradigm with enhanced privacy. Its primary goal is learning a global model that offers good performance for the participants as many as possible. The technology is rapidly advancing with many unsolved challenges, among which statistical heterogeneity (i.e., non-IID) and communication efficiency are two critical ones that hinder the development of federated learning. In this work, we propose LotteryFL -- a personalized and communication-efficient federated learning framework via exploiting the Lottery Ticket hypothesis. In LotteryFL, each client learns a lottery ticket network (i.e., a subnetwork of the base model) by applying the Lottery Ticket hypothesis, and only these lottery networks will be communicated between the server and clients. Rather than learning a shared global model in classic federated learning, each client learns a personalized model via LotteryFL; the communication cost can be significantly reduced due to the compact size of lottery networks. To support the training and evaluation of our framework, we construct non-IID datasets based on MNIST, CIFAR-10 and EMNIST by taking feature distribution skew, label distribution skew and quantity skew into consideration. Experiments on these non-IID datasets demonstrate that LotteryFL significantly outperforms existing solutions in terms of personalization and communication cost.
Node classification and graph classification are two basic graph analytics tools. Node classification aims to predict a label for each node in a graph, while graph classification aims to predict a label for the entire graph. Existing studies on graph neural networks (GNNs) in adversarial settings mainly focused on node classification, leaving GNN based graph classification largely unexplored. We aim to bridge this gap in this work. Specifically, we propose a subgraph based backdoor attack to GNN based graph classification. In our backdoor attack, a GNN classifier predicts an attacker-chosen target label for a testing graph once the attacker injects a predefined subgraph to the testing graph. Our empirical results on three real-world graph datasets show that our backdoor attacks are effective with small impact on a GNN's prediction accuracy for clean testing graphs. We generalize a state-of-the-art randomized smoothing based certified defense to defend against our backdoor attacks. Our empirical results show that the defense is ineffective in some cases, highlighting the needs of new defenses for our backdoor attacks.
We consider the blackbox transfer-based targeted adversarial attack threat model in the realm of deep neural network (DNN) image classifiers. Rather than focusing on crossing decision boundaries at the output layer of the source model, our method perturbs representations throughout the extracted feature hierarchy to resemble other classes. We design a flexible attack framework that allows for multi-layer perturbations and demonstrates state-of-the-art targeted transfer performance between ImageNet DNNs. We also show the superiority of our feature space methods under a relaxation of the common assumption that the source and target models are trained on the same dataset and label space, in some instances achieving a $10\times$ increase in targeted success rate relative to other blackbox transfer methods. Finally, we analyze why the proposed methods outperform existing attack strategies and show an extension of the method in the case when limited queries to the blackbox model are allowed.
Backdoor attack is a severe security threat to deep neural networks (DNNs). We envision that, like adversarial examples, there will be a cat-and-mouse game for backdoor attacks, i.e., new empirical defenses are developed to defend against backdoor attacks but they are soon broken by strong adaptive backdoor attacks. To prevent such cat-and-mouse game, we take the first step towards certified defenses against backdoor attacks. Specifically, in this work, we study the feasibility and effectiveness of certifying robustness against backdoor attacks using a recent technique called randomized smoothing. Randomized smoothing was originally developed to certify robustness against adversarial examples. We generalize randomized smoothing to defend against backdoor attacks. Our results show the theoretical feasibility of using randomized smoothing to certify robustness against backdoor attacks. However, we also find that existing randomized smoothing methods have limited effectiveness at defending against backdoor attacks, which highlight the needs of new theory and methods to certify robustness against backdoor attacks.
Community detection plays a key role in understanding graph structure. However, several recent studies showed that community detection is vulnerable to adversarial structural perturbation. In particular, via adding or removing a small number of carefully selected edges in a graph, an attacker can manipulate the detected communities. However, to the best of our knowledge, there are no studies on certifying robustness of community detection against such adversarial structural perturbation. In this work, we aim to bridge this gap. Specifically, we develop the first certified robustness guarantee of community detection against adversarial structural perturbation. Given an arbitrary community detection method, we build a new smoothed community detection method via randomly perturbing the graph structure. We theoretically show that the smoothed community detection method provably groups a given arbitrary set of nodes into the same community (or different communities) when the number of edges added/removed by an attacker is bounded. Moreover, we show that our certified robustness is tight. We also empirically evaluate our method on multiple real-world graphs with ground truth communities.
It is well-known that classifiers are vulnerable to adversarial perturbations. To defend against adversarial perturbations, various certified robustness results have been derived. However, existing certified robustnesses are limited to top-1 predictions. In many real-world applications, top-$k$ predictions are more relevant. In this work, we aim to derive certified robustness for top-$k$ predictions. In particular, our certified robustness is based on randomized smoothing, which turns any classifier to a new classifier via adding noise to an input example. We adopt randomized smoothing because it is scalable to large-scale neural networks and applicable to any classifier. We derive a tight robustness in $\ell_2$ norm for top-$k$ predictions when using randomized smoothing with Gaussian noise. We find that generalizing the certified robustness from top-1 to top-$k$ predictions faces significant technical challenges. We also empirically evaluate our method on CIFAR10 and ImageNet. For example, our method can obtain an ImageNet classifier with a certified top-5 accuracy of 62.8\% when the $\ell_2$-norms of the adversarial perturbations are less than 0.5 (=127/255). Our code is publicly available at: \url{https://github.com/jjy1994/Certify_Topk}.