SAND: A Self-supervised and Adaptive NAS-Driven Framework for Hardware Trojan Detection

2025-10-24

AI Summary

Hardware Trojan (HT) detection faces challenges including heavy reliance on manual feature engineering and poor model generalizability. To address these, this paper proposes an end-to-end detection framework synergistically driven by self-supervised learning and neural architecture search (NAS). First, self-supervised pretraining enables unsupervised, automated feature extraction from raw circuit representations. Subsequently, NAS dynamically discovers the optimal classifier architecture, which is then fine-tuned on downstream HT detection tasks to enhance task-specific adaptability. By eliminating handcrafted feature design, the method significantly improves robustness against diverse, stealthy, and previously unseen HTs. Evaluated on multiple benchmark circuit suites, the proposed approach achieves up to 18.3% higher detection accuracy than state-of-the-art methods, demonstrating superior generalization and detection capability.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Learning to Search
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

The globalized semiconductor supply chain has made Hardware Trojans (HT) a significant security threat to embedded systems, necessitating the design of efficient and adaptable detection mechanisms. Despite promising machine learning-based HT detection techniques in the literature, they suffer from ad hoc feature selection and the lack of adaptivity, all of which hinder their effectiveness across diverse HT attacks. In this paper, we propose SAND, a selfsupervised and adaptive NAS-driven framework for efficient HT detection. Specifically, this paper makes three key contributions. (1) We leverage self-supervised learning (SSL) to enable automated feature extraction, eliminating the dependency on manually engineered features. (2) SAND integrates neural architecture search (NAS) to dynamically optimize the downstream classifier, allowing for seamless adaptation to unseen benchmarks with minimal fine-tuning. (3) Experimental results show that SAND achieves a significant improvement in detection accuracy (up to 18.3%) over state-of-the-art methods, exhibits high resilience against evasive Trojans, and demonstrates strong generalization.

Problem

• Automating feature extraction for Hardware Trojan detection using self-supervised learning
• Dynamically optimizing classifiers via neural architecture search for adaptability
• Improving detection accuracy and resilience against diverse Trojan attacks

Innovation

• Self-supervised learning automates hardware Trojan feature extraction
• Neural architecture search dynamically optimizes detection classifier
• Framework achieves high accuracy with minimal fine-tuning adaptation