DETERRENT: Detecting Trojans using Reinforcement Learning
This addresses a critical security threat in hardware design by improving detection efficiency, though it is an incremental advancement in applying RL to a known bottleneck.
The paper tackled the problem of detecting hardware Trojans in integrated circuits by designing a reinforcement learning agent to generate minimal test patterns, achieving a 169x reduction in patterns while maintaining 95.75% coverage compared to state-of-the-art methods.
Insertion of hardware Trojans (HTs) in integrated circuits is a pernicious threat. Since HTs are activated under rare trigger conditions, detecting them using random logic simulations is infeasible. In this work, we design a reinforcement learning (RL) agent that circumvents the exponential search space and returns a minimal set of patterns that is most likely to detect HTs. Experimental results on a variety of benchmarks demonstrate the efficacy and scalability of our RL agent, which obtains a significant reduction ($169\times$) in the number of test patterns required while maintaining or improving coverage ($95.75\%$) compared to the state-of-the-art techniques.