Logic locking protects the integrity of hardware designs throughout the integrated circuit supply chain. However, recent machine learning (ML)-based attacks have challenged its fundamental security, initiating the requirement for the design of learning-resilient locking policies. A promising ML-resilient locking mechanism hides within multiplexer-based locking. Nevertheless, recent attacks have successfully breached these state-of-the-art locking schemes, making it ever more complex to manually design policies that are resilient to all existing attacks. In this project, for the first time, we propose the automatic design exploration of logic locking with evolutionary computation (EC) -- a set of versatile black-box optimization heuristics inspired by evolutionary mechanisms. The project will evaluate the performance of EC-designed logic locking against various types of attacks, starting with the latest ML-based link prediction. Additionally, the project will provide guidelines and best practices for using EC-based logic locking in practical applications.

翻译：逻辑锁定技术用于保护集成电路供应链中硬件设计的完整性。然而，近年来基于机器学习（ML）的攻击对其基本安全性提出了挑战，促使我们需要设计具有抗学习能力的锁定策略。一种具有前景的抗ML锁定机制隐藏在基于多路选择器的锁定方案中。尽管如此，近期攻击已成功突破这些最先进的锁定方案，使得手动设计能够抵御所有现有攻击的策略变得愈发复杂。在本项目中，我们首次提出利用进化计算（EC）——一组受进化机制启发的通用黑箱优化启发式算法——对逻辑锁定进行自动化设计探索。该项目将评估经EC设计的逻辑锁定针对各类攻击（从最新的基于ML的链路预测攻击开始）的性能表现。此外，项目还将为在实际应用中使用基于EC的逻辑锁定提供指导方针与最佳实践。