• Monday | October 27, 2025 | 10:00 - 11:30

    The ever-increasing complexity of microprocessors has resulted in several potent security threats in recent years. These vulnerabilities in the hardware arising from unchecked performance optimizations have been exploited through various ways, such as micro-architectural attacks, fault injections, memory corruption, and other forms of information leakage. Post tape-out, hardware vulnerabilities are typically mitigated using software updates or hardware recall, which result in unacceptably high performance or economic overheads, respectively. Thus, there is a pressing need to uncover these vulnerabilities during the hardware design phase. Integrating such approaches can improve the overall security, reliability, and economic viability of microprocessors. In this tutorial, we introduce hardware vulnerabilities and state-of-the-art techniques to uncover these vulnerabilities at design time, leveraging hardware fuzzing, AI, and formal verification. Tutorial participants will gain an understanding of the fundamentals of hardware vulnerabilities, their origins, and detection approaches. We will present some of the potent Common Weakness Enumerations (CWEs) we have exposed in popular microprocessors. With our assistance, participants will get a real-world demonstration of the hardware fuzzing techniques used to detect these vulnerabilities and pinpoint their location in hardware design. We will explore the recent advances in hardware fuzzing using AI techniques and formal verification. We will use concrete, hands-on examples to quantitatively analyze the potential of these techniques for hardware security and the open challenges in the domain.

    • Sadeghi_560x720_150dpi_415x0_415x0.jpg
      Ahmad-Reza Sadeghi

      Technical University of Darmstadt Darmstadt, Hesse, Germany

    • JV Rajendran.png
      JV Rajendran

      Texas A&M University

    • Nikhilesh Singh.png
      Nikhilesh Singh

      TU Darmstadt

    • Lichao Wu.png
      Lichao Wu

      TU Darmstadt

    • Huimin Li.png
      Huimin Li

      TU Darmstadt

    • Chen Chen.png
      Chen Chen

      Texas A&M University

    • Mohamadreza Rostami.png
      Mohamadreza Rostami

      TU Darmstadt

  • Monday | October 27, 2025 | 13:00 - 14:30

    This tutorial explores the intersection of quantum machine learning (QML) and the design of quantum architectures, highlighting foundational techniques that pave the way for future quantum-enabled design automation (Quantum EDA). Rather than optimizing classical semiconductor flows, we focus on how QML, quantum reinforcement learning (QRL), and quantum architecture search (QAS) can be applied to discover, optimize, and innovate quantum circuits and variational architectures. Through interactive, hands-on sessions using open-source quantum simulators such as Qiskit and PennyLane, participants will engage with key methodologies for quantum model design and architectural optimization. This session is aimed at future explorers — those who seek to rethink design automation in the quantum era — and no prior knowledge of traditional EDA is required.

    • Samuel Yen-Chi Chen.jpg
      Samuel Yen-Chi Chen

      Wells Fargo

    • Zhiding Liang.png
      Zhiding Liang

      The Chinese University of Hong Kong

    • Kuan-Chen Chen.png
      Kuan-Chen Chen

      Imperial College London