Regression-aware Continual Learning for Android Malware Detection

• URL: http://arxiv.org/abs/2507.18313v1
• Date: Thu, 24 Jul 2025 11:31:23 GMT
• Title: Regression-aware Continual Learning for Android Malware Detection
• Authors: Daniele Ghiani, Daniele Angioni, Giorgio Piras, Angelo Sotgiu, Luca Minnei, Srishti Gupta, Maura Pintor, Fabio Roli, Battista Biggio,
• Abstract summary: Malware evolves rapidly, forcing machine learning (ML)-based detectors to adapt continuously.<n>Continual learning (CL) has emerged as a scalable alternative, enabling incremental updates without full data access.<n>But security regression captures harmful prediction changes at the sample level, such as a malware sample that was once correctly detected but evades detection after a model update.<n>We formalize and quantify security regression in CL-based malware detectors and propose a regression-aware penalty to mitigate it.
• Score: 9.695692033183485
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Malware evolves rapidly, forcing machine learning (ML)-based detectors to adapt continuously. With antivirus vendors processing hundreds of thousands of new samples daily, datasets can grow to billions of examples, making full retraining impractical. Continual learning (CL) has emerged as a scalable alternative, enabling incremental updates without full data access while mitigating catastrophic forgetting. In this work, we analyze a critical yet overlooked issue in this context: security regression. Unlike forgetting, which manifests as a general performance drop on previously seen data, security regression captures harmful prediction changes at the sample level, such as a malware sample that was once correctly detected but evades detection after a model update. Although often overlooked, regressions pose serious risks in security-critical applications, as the silent reintroduction of previously detected threats in the system may undermine users' trust in the whole updating process. To address this issue, we formalize and quantify security regression in CL-based malware detectors and propose a regression-aware penalty to mitigate it. Specifically, we adapt Positive Congruent Training (PCT) to the CL setting, preserving prior predictive behavior in a model-agnostic manner. Experiments on the ELSA, Tesseract, and AZ-Class datasets show that our method effectively reduces regression across different CL scenarios while maintaining strong detection performance over time.

Related papers

• A Sysmon Incremental Learning System for Ransomware Analysis and Detection [1.495391051525033]
  In the face of increasing cyber threats, particularly ransomware attacks, there is a pressing need for advanced detection and analysis systems.<n>Most of these proposals leverage non-incremental learning approaches that require the underlying models to be updated from scratch to detect new ransomware.<n>This approach is problematic because it leaves sensitive data vulnerable to attack during retraining, as newly emerging ransomware strains may go undetected until the model is updated.<n>We present the Sysmon Incremental Learning System for Analysis and Detection (SILRAD), which enables continuous updates to the underlying model and effectively closes the training gap.
  arXiv  Detail & Related papers  (2025-01-02T06:22:58Z)
• Revisiting Concept Drift in Windows Malware Detection: Adaptation to Real Drifted Malware with Minimal Samples [10.352741619176383]
  We propose a new technique for detecting and classifying drifted malware.<n>It learns drift-invariant features in malware control flow graphs by leveraging graph neural networks with adversarial domain adaptation.<n>Our approach significantly improves drifted malware detection on publicly available benchmarks and real-world malware databases reported daily by security companies.
  arXiv  Detail & Related papers  (2024-07-18T22:06:20Z)
• Updating Windows Malware Detectors: Balancing Robustness and Regression against Adversarial EXEmples [8.911277142080934]
  We study the trade-off between accuracy and regression when updating Windows malware detectors.<n>We propose EXE-scanner, a plugin that can be chained to existing detectors to promptly stop EXEmples without causing regression.
  arXiv  Detail & Related papers  (2024-05-04T12:12:23Z)
• Client-side Gradient Inversion Against Federated Learning from Poisoning [59.74484221875662]
  Federated Learning (FL) enables distributed participants to train a global model without sharing data directly to a central server. Recent studies have revealed that FL is vulnerable to gradient inversion attack (GIA), which aims to reconstruct the original training samples. We propose Client-side poisoning Gradient Inversion (CGI), which is a novel attack method that can be launched from clients.
  arXiv  Detail & Related papers  (2023-09-14T03:48:27Z)
• Exploring Model Dynamics for Accumulative Poisoning Discovery [62.08553134316483]
  We propose a novel information measure, namely, Memorization Discrepancy, to explore the defense via the model-level information. By implicitly transferring the changes in the data manipulation to that in the model outputs, Memorization Discrepancy can discover the imperceptible poison samples. We thoroughly explore its properties and propose Discrepancy-aware Sample Correction (DSC) to defend against accumulative poisoning attacks.
  arXiv  Detail & Related papers  (2023-06-06T14:45:24Z)
• Fast & Furious: Modelling Malware Detection as Evolving Data Streams [6.6892028759947175]
  Malware is a major threat to computer systems and imposes many challenges to cyber security. In this work, we evaluate the impact of concept drift on malware classifiers for two Android datasets.
  arXiv  Detail & Related papers  (2022-05-24T18:43:40Z)
• Do Gradient Inversion Attacks Make Federated Learning Unsafe? [70.0231254112197]
  Federated learning (FL) allows the collaborative training of AI models without needing to share raw data. Recent works on the inversion of deep neural networks from model gradients raised concerns about the security of FL in preventing the leakage of training data. In this work, we show that these attacks presented in the literature are impractical in real FL use-cases and provide a new baseline attack.
  arXiv  Detail & Related papers  (2022-02-14T18:33:12Z)
• Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In NLP Model Updates [68.09049111171862]
  This work focuses on quantifying, reducing and analyzing regression errors in the NLP model updates. We formulate the regression-free model updates into a constrained optimization problem. We empirically analyze how model ensemble reduces regression.
  arXiv  Detail & Related papers  (2021-05-07T03:33:00Z)
• Continual Learning for Fake Audio Detection [62.54860236190694]
  This paper proposes detecting fake without forgetting, a continual-learning-based method, to make the model learn new spoofing attacks incrementally. Experiments are conducted on the ASVspoof 2019 dataset.
  arXiv  Detail & Related papers  (2021-04-15T07:57:05Z)
• Detection as Regression: Certified Object Detection by Median Smoothing [50.89591634725045]
  This work is motivated by recent progress on certified classification by randomized smoothing. We obtain the first model-agnostic, training-free, and certified defense for object detection against $ell$-bounded attacks.
  arXiv  Detail & Related papers  (2020-07-07T18:40:19Z)
• MDEA: Malware Detection with Evolutionary Adversarial Learning [16.8615211682877]
  MDEA, an Adversarial Malware Detection model uses evolutionary optimization to create attack samples to make the network robust against evasion attacks. By retraining the model with the evolved malware samples, its performance improves a significant margin.
  arXiv  Detail & Related papers  (2020-02-09T09:59:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.