Toward Realistic Adversarial Attacks in IDS: A Novel Feasibility Metric for Transferability

2025-04-11

AI Summary

This paper addresses the practical feasibility of transferable adversarial attacks against Intrusion Detection Systems (IDS) under limited source/target model similarity—a critical yet underexplored challenge. We propose the Transferability Feasibility Score (TFS), the first quantitative metric for evaluating attack transfer success probability. TFS jointly models three dimensions: feature alignment, architectural similarity, and data distribution overlap, thereby bridging the gap between theoretical transferability and empirically observed attack success rates. Through extensive empirical validation within a multidimensional analytical framework, we demonstrate that TFS achieves strong correlation with actual attack success (Pearson’s *r* > 0.92), substantially enhancing predictability of transfer attacks. Moreover, TFS provides actionable, interpretable quantification to guide the design and evaluation of robust defensive mechanisms against such attacks.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Multiagent Systems: Adversarial Agents
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Transferability-based adversarial attacks exploit the ability of adversarial examples, crafted to deceive a specific source Intrusion Detection System (IDS) model, to also mislead a target IDS model without requiring access to the training data or any internal model parameters. These attacks exploit common vulnerabilities in machine learning models to bypass security measures and compromise systems. Although the transferability concept has been widely studied, its practical feasibility remains limited due to assumptions of high similarity between source and target models. This paper analyzes the core factors that contribute to transferability, including feature alignment, model architectural similarity, and overlap in the data distributions that each IDS examines. We propose a novel metric, the Transferability Feasibility Score (TFS), to assess the feasibility and reliability of such attacks based on these factors. Through experimental evidence, we demonstrate that TFS and actual attack success rates are highly correlated, addressing the gap between theoretical understanding and real-world impact. Our findings provide needed guidance for designing more realistic transferable adversarial attacks, developing robust defenses, and ultimately improving the security of machine learning-based IDS in critical systems.

Problem

• Assessing feasibility of transferability-based adversarial attacks on IDS
• Proposing Transferability Feasibility Score to measure attack reliability
• Bridging gap between theoretical transferability and real-world impact

Innovation

• Proposes Transferability Feasibility Score (TFS)
• Analyzes feature alignment and model similarity
• Links TFS to real-world attack success rates