mlsanityData to 7 Oct 2026

Deceiving End-to-End Deep Learning Malware Detectors using Adversarial Examples

arXiv 1802.04528 ↗First posted 13 Feb 2018cs.LG cs.CRcited by 19

Not on the map. Why?

Abstract (arXiv)

In recent years, deep learning has shown performance breakthroughs in many applications, such as image detection, image segmentation, pose estimation, and speech recognition. However, this comes with a major concern: deep networks have been found to be vulnerable to adversarial examples. Adversarial examples are slightly modified inputs that are intentionally designed to cause a misclassification by the model. In the domains of images and speech, the modifications are so small that they are not seen or heard by humans, but nevertheless greatly affect the classification of the model. Deep learning models have been successfully applied to malware detection. In this domain, generating adversarial examples is not straightforward, as small modifications to the bytes of the file could lead to significant changes in its functionality and validity. We introduce a novel loss function for generating adversarial examples specifically tailored for discrete input sets, such as executable bytes. We modify malicious binaries so that they would be detected as benign, while preserving their original functionality, by injecting a small sequence of bytes (payload) in the binary file. We applied this approach to an end-to-end convolutional deep learning malware detection model and show a high rate of detection evasion. Moreover, we show that our generated payload is robust enough to be transferable within different locations of the same file and across different files, and that its entropy is low and similar to that of benign data sections.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). First posted on arXiv 13 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 2; 2023: 6; 2024: 4; 2025: 0; 2026: 1.

19 papers.

Sort:

  1. 2021Adversarial Attacks against Windows PE Malware Detection: A Survey of the State-of-the-Artcited by 18
  2. 2023RS-Del: Edit Distance Robustness Certificates for Sequence Classifiers via Randomized Deletioncited by 17
  3. 2021EvadeDroid: A Practical Evasion Attack on Machine Learning for Black-box Android Malware Detectioncited by 13
  4. 2023A Survey on Malware Detection with Graph Representation Learningcited by 11
  5. 2021Explanation-Guided Diagnosis of Machine Learning Evasion Attackscited by 6
  6. 2023A Comparison of Adversarial Learning Techniques for Malware Detectioncited by 5
  7. 2024A Robust Defense against Adversarial Attacks on Deep Learning-based Malware Detectors via (De)Randomized Smoothingcited by 4
  8. 2021Mal2GCN: A Robust Malware Detection Approach Using Deep Graph Convolutional Networks With Non-Negative Weightscited by 3
  9. 2023DRSM: De-Randomized Smoothing on Malware Classifier Providing Certified Robustnesscited by 3
  10. 2023Creating Valid Adversarial Examples of Malwarecited by 3
  11. 2022Practical Attacks on Machine Learning: A Case Study on Adversarial Windows Malwarecited by 2
  12. 2023Combining Generators of Adversarial Malware Examples to Increase Evasion Ratecited by 2
7 more
  1. 2022Artificial Intelligence for Cybersecurity: Threats, Attacks and Mitigationcited by 1
  2. 2024CARE: Ensemble Adversarial Robustness Evaluation Against Adaptive Attackers for Security Applicationscited by 1
  3. 2021ML-based IoT Malware Detection Under Adversarial Settings: A Systematic Evaluationcited by 0
  4. 2021A Comparison of State-of-the-Art Techniques for Generating Adversarial Malware Binariescited by 0
  5. 2024Towards Robust Domain Generation Algorithm Classificationcited by 0
  6. 2024Certified Adversarial Robustness of Machine Learning-based Malware Detectors via (De)Randomized Smoothingcited by 0
  7. 2026Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbationscited by 0