EDAML 2022 Invited Speaker 7: Analog and Digital Circuit and Layout Optimization using Machine Learning

Machine learning (ML) has opened new doors for a wide range of techniques that can increase the efficiency of the design cycle — ranging from methods that traditionally require manual input from expert designers to methods that perform fast analyses and optimizations. This talk will overview some of our recent research on applying ML methods to problems in electronic design automation (EDA). We will describe methods that are used for circuit recognition, power grid analysis and optimization, and realistic benchmark generation, describing the ability of ML to bring new capabilities to EDA that were previously unavailable.