Strong evasive backdoor attacks and an ensemble defense

Summary

Researchers present an ensemble defense that combines several backdoor detectors for deep neural network classifiers, so detection does not depend on a single backdoor mechanism. The ensemble also performs backdoor inversion, which indicates the nature of a detected attack. The paper also employs mixed clean/dirty-label backdoor poisoning, an X-to-X attack the authors describe as more surgical, evasive, and harder to detect than traditional dirty-label attacks.

Topics

• Adversarial machine learning

Related items

• LowGNN-Guided Selection of Benign-like Anomalies for Backdoor Attacks Against Network Intrusion Detection SystemsSimilar attack · OpenAlex (peer-reviewed AI security)
• InfoRolling the cyber dice with open-source and open-weight AI modelsSimilar attack · CSO Online
• InfoAEGIS: Real-Time Latent-Space Backdoor Detection for Dependable and Secure Small Language Model InferenceSimilar attack · OpenAlex (peer-reviewed AI security)
• InfoHFL-Deflect: Defence framework against adaptive coordinated poisoning attacks in hierarchical federated learningSimilar attack · Elsevier Security Journals
• InfoSecurity Threat Detection and Defense Theory of Intelligent Operation and Maintenance System for Rail Transit Based on Adversarial Machine LearningSimilar attack · OpenAlex (peer-reviewed AI security)