Scholarlink

<p>Trojan attacks are sophisticated training-time attacks on neural networks that embed backdoor triggers that force the network to produce a specific output on any input that includes the trigger. With the increasing relevance of deep networks that are too large to train with personal resources and that are trained on data too large to thoroughly audit, these training-time attacks pose a significant risk. In this work, we connect trojan attacks to Neural Collapse, a phenomenon wherein the final feature representations of over-parameterized neural networks converge to a simple geometric structure. We provide experimental evidence that trojan attacks disrupt this convergence for a variety of datasets and architectures. We then use this disruption to design a lightweight, broadly generalizable mechanism for cleansing trojan attacks from a wide variety of different network architectures and experimentally demonstrate its efficacy.</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

Trojan cleansing with neural collapse

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月20日

查看全文
  • Xihe Gu,
  • Greg Fields,
  • Yaman Jandali,
  • Farinaz Koushanfar,
  • Tara Javidi

摘要

Trojan attacks are sophisticated training-time attacks on neural networks that embed backdoor triggers that force the network to produce a specific output on any input that includes the trigger. With the increasing relevance of deep networks that are too large to train with personal resources and that are trained on data too large to thoroughly audit, these training-time attacks pose a significant risk. In this work, we connect trojan attacks to Neural Collapse, a phenomenon wherein the final feature representations of over-parameterized neural networks converge to a simple geometric structure. We provide experimental evidence that trojan attacks disrupt this convergence for a variety of datasets and architectures. We then use this disruption to design a lightweight, broadly generalizable mechanism for cleansing trojan attacks from a wide variety of different network architectures and experimentally demonstrate its efficacy.