Machine learning and systems for the next frontier in formal verification

Summary form only given. This tutorial covers basics of machine learning, systems and infrastructure considerations for performing machine learning at scale, and applications of machine learning to improve formal verification performance and usability. It starts with blackbox classifier training with gradient descent, and proceeds on to deep network training and simple convolutional neural networks. Next, it discusses how machine learning can be performed at scale, overcoming the performance and throughput limitations of traditional compute and storage systems. Finally, the tutorial describes several ways in which machine learning can be applied for improving formal tools performance and enhancing debug capabilities.