An AI Architecture with the Capability to Classify and Explain Hardware Trojans
This addresses the need for interpretable AI in hardware security, but appears incremental as it builds on existing detection features.
The paper tackles the problem of hardware trojan detection methods lacking explainability by introducing an explainable methodology and architecture based on existing detection features, providing results for explaining digital hardware trojans using trust-hub benchmarks.
Hardware trojan detection methods, based on machine learning (ML) techniques, mainly identify suspected circuits but lack the ability to explain how the decision was arrived at. An explainable methodology and architecture is introduced based on the existing hardware trojan detection features. Results are provided for explaining digital hardware trojans within a netlist using trust-hub trojan benchmarks.