Malware remains a big threat to cyber security, calling for machine learning based malware detection. While promising, such detectors are known to be vulnerable to evasion attacks. Ensemble learning typically facilitates countermeasures, while attackers can leverage this technique to improve attack effectiveness as well. This motivates us to investigate which kind of robustness the ensemble defense or effectiveness the ensemble attack can achieve, particularly when they combat with each other. We thus propose a new attack approach, named mixture of attacks, by rendering attackers capable of multiple generative methods and multiple manipulation sets, to perturb a malware example without ruining its malicious functionality. This naturally leads to a new instantiation of adversarial training, which is further geared to enhancing the ensemble of deep neural networks. We evaluate defenses using Android malware detectors against 26 different attacks upon two practical datasets. Experimental results show that the new adversarial training significantly enhances the robustness of deep neural networks against a wide range of attacks, ensemble methods promote the robustness when base classifiers are robust enough, and yet ensemble attacks can evade the enhanced malware detectors effectively, even notably downgrading the VirusTotal service.
13 January 2021 in IEEE Transactions on Network Science and Engineering
Machine learning-based malware detection is known to be vulnerable to adversarial evasion attacks. The state-of-the-art is that there are no effective defenses against these attacks. As a response to the adversarial malware classification challenge organized by the MIT Lincoln Lab and associated with the AAAI-19 Workshop on Artificial Intelligence for Cyber Security (AICS'2019), we propose six guiding principles to enhance the robustness of deep neural networks. Some of these principles have been scattered in the literature, but the others are introduced in this paper for the first time. Under the guidance of these six principles, we propose a defense framework to enhance the robustness of deep neural networks against adversarial malware evasion attacks. By conducting experiments with the Drebin Android malware dataset, we show that the framework can achieve a 98.49% accuracy (on average) against grey-box attacks, where the attacker knows some information about the defense and the defender knows some information about the attack, and an 89.14% accuracy (on average) against the more capable white-box attacks, where the attacker knows everything about the defense and the defender knows some information about the attack. The framework wins the AICS'2019 challenge by achieving a 76.02% accuracy, where neither the attacker (i.e., the challenge organizer) knows the framework or defense nor we (the defender) know the attacks. This gap highlights the importance of knowing about the attack.
7 April 2023 in IEEE Transactions on Dependable and Secure Computing
Machine Learning (ML) techniques can facilitate the automation of mal icious soft ware (malware for short) detection, but suffer from evasion attacks. Many studies counter such attacks in heuristic manners, lacking theoretical guarantees and defense effectiveness. In this paper, we propose a new adversarial training framework, termed P rincipled A dversarial Malware D etection (PAD), which offers convergence guarantees for robust optimization methods. PAD lays on a learnable convex measurement that quantifies distribution-wise discrete perturbations to protect malware detectors from adversaries, whereby for smooth detectors, adversarial training can be performed with theoretical treatments. To promote defense effectiveness, we propose a new mixture of attacks to instantiate PAD to enhance deep neural network-based measurements and malware detectors. Experimental results on two Android malware datasets demonstrate: (i) the proposed method significantly outperforms the state-of-the-art defenses; (ii) it can harden ML-based malware detection against 27 evasion attacks with detection accuracies greater than 83.45%, at the price of suffering an accuracy decrease smaller than 2.16% in the absence of attacks; (iii) it matches or outperforms many anti-malware scanners in VirusTotal against realistic adversarial malware.
Graph Neural Networks (GNNs) have been proven to be useful for learning graph-based knowledge. However, one of the drawbacks of GNN techniques is that they may get stuck in the problem of over-squashing. Recent studies attribute to the message passing paradigm that it may amplify some specific local relations and distort long-range information under a certain GNN. To alleviate such phenomena, we propose a novel and general GNN framework, dubbed MC-GNN, which introduces the multi-channel neural architecture to learn and fuse multi-view graph-based information. The purpose of MC-GNN is to extract distinct channel-based graph features and adaptively adjust the importance of the features. To this end, we use the Hilbert-Schmidt Independence Criterion (HSIC) to enlarge the disparity between the embeddings encoded by each channel and follow an attention mechanism to fuse the embeddings with adaptive weight adjustment. MC-GNN can apply multiple GNN backbones, which provides a solution for learning structural relations from a multi-view perspective. Experimental results demonstrate that the proposed MC-GNN is superior to the compared state-of-the-art GNN methods.
Malware continues to be a major cyber threat, despite the tremendous effort that has been made to combat them. The number of malware in the wild steadily increases over time, meaning that we must resort to automated defense techniques. This naturally calls for machine learning based malware detection. However, machine learning is known to be vulnerable to adversarial evasion attacks that manipulate a small number of features to make classifiers wrongly recognize a malware sample as a benign one. The state-of-the-art is that there are no effective countermeasures against these attacks. Inspired by the AICS'2019 Challenge, we systematize a number of principles for enhancing the robustness of neural networks against adversarial malware evasion attacks. Some of these principles have been scattered in the literature, but others are proposed in this paper for the first time. Under the guidance of these principles, we propose a framework and an accompanying training algorithm, which are then applied to the AICS'2019 challenge. Our experimental results have been submitted to the challenge organizer for evaluation.