Scholarlink

<p>Deep learning-based malware detectors remain highly vulnerable to adversarial attacks. While individual defense mechanisms such as denoising autoencoders, adversarial training, and generative networks have been explored separately, their orchestrated integration into a unified, multi-layer defense pipeline remains an open challenge. This paper introduces DefendMal, a novel framework that synergistically combines Denoise Autoencoder with Sequence Squeezing (DA-SS), a Context-aware Adversarial Generator (CAG-AdvGAN), Projected Gradient Descent (PGD) adversarial training, and a Positive–Negative Detector with Variational Autoencoder (PNDetector-VAE) to enhance robustness against evolving adversarial threats. Unlike prior works that focus on isolated defenses, DefendMal employs a cascaded defense strategy&#xa0;that sequentially preprocesses inputs, generates adaptive attacks, hardens the model, and detects poisoned samples. We evaluate DefendMal on a curated malware-inspired adversarial dataset derived from image-based perturbations, serving as a reproducible proxy for malware feature space. Our results demonstrate that DefendMal achieves 97% accuracy and 98% precision under adversarial conditions, significantly outperforming state-of-the-art baselines. We further validate statistical reliability through confidence intervals derived from multiple experimental runs and ablation studies confirm the contribution of each component.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing robustness of deep learning-based malware detection against adversarial attacks

  • 期刊论文

  • 发表日期：

    2026年8月2日

查看全文
  • Dennis Benedict Crasta,
  • Vikas Kumar

摘要

Deep learning-based malware detectors remain highly vulnerable to adversarial attacks. While individual defense mechanisms such as denoising autoencoders, adversarial training, and generative networks have been explored separately, their orchestrated integration into a unified, multi-layer defense pipeline remains an open challenge. This paper introduces DefendMal, a novel framework that synergistically combines Denoise Autoencoder with Sequence Squeezing (DA-SS), a Context-aware Adversarial Generator (CAG-AdvGAN), Projected Gradient Descent (PGD) adversarial training, and a Positive–Negative Detector with Variational Autoencoder (PNDetector-VAE) to enhance robustness against evolving adversarial threats. Unlike prior works that focus on isolated defenses, DefendMal employs a cascaded defense strategy that sequentially preprocesses inputs, generates adaptive attacks, hardens the model, and detects poisoned samples. We evaluate DefendMal on a curated malware-inspired adversarial dataset derived from image-based perturbations, serving as a reproducible proxy for malware feature space. Our results demonstrate that DefendMal achieves 97% accuracy and 98% precision under adversarial conditions, significantly outperforming state-of-the-art baselines. We further validate statistical reliability through confidence intervals derived from multiple experimental runs and ablation studies confirm the contribution of each component.