Chris Bennett (University of Bristol )

Microelectronic design verification remains a critical bottleneck in device development, traditionally mitigated by expanding verification teams and computational resources. Since the late 1990s, machine learning (ML) has been proposed to enhance verification efficiency, yet many techniques have not achieved mainstream adoption, resulting in a wealth of research material but few practical steps to navigate the options.

This presentation takes the perspective of verification and ML practitioners. It examines research into the application of ML in test-based techniques for functional verification of microelectronic designs, providing a starting point for those new to this interdisciplinary field. Based on recent work at the University of Bristol, analysis and resources are presented to support practitioners looking to apply ML techniques to their verification and understand the pros and cons of different techniques relative to the problem at hand. Practical next steps from both industry and academia are also discussed to enable the transition of machine learning from research to real-world use.

Key Points:

• A wide variety of ML techniques and applications can reduce the resources needed to verify a design, but it remains challenging to identify which are the most suitable.
• The growing interest in and quality of open-source designs creates opportunities for establishing common benchmarks to compare techniques and for the industry to guide the research direction.
• Software verification may offer solutions to hardware verification challenges.

Chris Bennett is a senior research associate at the University of Bristol, developing machine learning techniques for test-based verification. The work forms part of a consortium of UK industry and academia developing the next generation of microprocessors for use in aerospace and other harsh environments. A chartered engineer with a background in systems engineering for automotive, he worked at Jaguar Land Rover before transitioning into research seven years ago, completing a PhD in Robotics and Autonomous Systems. He has previously worked on projects with Thales UK, examining the role of hybrid autonomy in multi-agent systems, and on the UKRI funded Trustworthy Autonomous Systems project, investigating how trust can be built in artificial intelligence and robotics through system engineering practices. His research interests include test-based verification, system engineering design practices, and multi-agent artificial intelligence.