mlsanityData to 7 Oct 2026

PAD: Towards Principled Adversarial Malware Detection Against Evasion Attacks

arXiv 2302.11328 ↗First posted 22 Feb 2023cs.CR cs.LG stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

Machine Learning (ML) techniques can facilitate the automation of malicious software (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 Principled Adversarial Malware Detection (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.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 40 of its 80 references that are in the map. First posted on arXiv 22 Feb 2023.

Builds on

40 of 80 references.

References · 40

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2017On Calibration of Modern Neural Networkscited by 4,860
  6. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  7. 2015Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)cited by 1,061
  8. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  9. 2018A Convergence Theory for Deep Learning via Over-Parameterizationcited by 784
  10. 2017Evasion Attacks against Machine Learning at Test Timecited by 686
  11. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 542
  12. 2016Input Convex Neural Networkscited by 407
28 more
  1. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 350
  2. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  3. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 276
  4. 2017On the (Statistical) Detection of Adversarial Examplescited by 180
  5. 2018Robustness via curvature regularization, and vice versacited by 174
  6. 2021On the Convergence and Robustness of Adversarial Trainingcited by 127
  7. 2017Malware Detection by Eating a Whole EXEcited by 114
  8. 2022LAS-AT: Adversarial Training with Learnable Attack Strategycited by 94
  9. 2018Technical Report: When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attackscited by 78
  10. 2018Adversarial Malware Binaries: Evading Deep Learning for Malware Detection in Executablescited by 60
  11. 2020Functionality-preserving Black-box Optimization of Adversarial Windows Malwarecited by 58
  12. 2017Yes, Machine Learning Can Be More Secure! A Case Study on Android Malware Detectioncited by 51
  13. 2019One-Class Convolutional Neural Networkcited by 37
  14. 2018Discrete Adversarial Attacks and Submodular Optimization with Applications to Text Classificationcited by 34
  15. 2021Evading Adversarial Example Detection Defenses with Orthogonal Projected Gradient Descentcited by 34
  16. 2020Adversarial Deep Ensemble: Evasion Attacks and Defenses for Malware Detectioncited by 33
  17. 2018Adversarial Deep Learning for Robust Detection of Binary Encoded Malwarecited by 32
  18. 2016MaMaDroid: Detecting Android Malware by Building Markov Chains of Behavioral Modelscited by 29
  19. 2020On the Generalization Properties of Adversarial Trainingcited by 23
  20. 2019DLA: Dense-Layer-Analysis for Adversarial Example Detectioncited by 20
  21. 2018Android HIV: A Study of Repackaging Malware for Evading Machine-Learning Detectioncited by 19
  22. 2019On Training Robust PDF Malware Classifierscited by 19
  23. 2017Automated Poisoning Attacks and Defenses in Malware Detection Systems: An Adversarial Machine Learning Approachcited by 18
  24. 2021Deep Learning for Android Malware Defenses: a Systematic Literature Reviewcited by 18
  25. 2020Arms Race in Adversarial Malware Detection: A Surveycited by 16
  26. 2020A Framework for Enhancing Deep Neural Networks Against Adversarial Malwarecited by 6
  27. 2018HashTran-DNN: A Framework for Enhancing Robustness of Deep Neural Networks against Adversarial Malware Samplescited by 4
  28. 2014An Evasion and Counter-Evasion Study in Malicious Websites Detectioncited by 3

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 1; 2025: 3.

5 papers.

  1. 2023MalPurifier: Enhancing Android Malware Detection with Adversarial Purification against Evasion Attackscited by 3
  2. 2025Defending against Adversarial Malware Attacks on ML-based Android Malware Detection Systemscited by 2
  3. 2024On the Effectiveness of Adversarial Training on Malware Classifierscited by 1
  4. 2025On the Security Risks of ML-based Malware Detection Systems: A Surveycited by 1
  5. 2025Evaluating the robustness of adversarial defenses in malware detection systemscited by 0