Session

DescriptionThis session delves into the application of machine learning across different stages of hardware design, from analog circuits, sequential circuits, physical design, and simulation. The first set of presentations focuses on leveraging AI for optimizing analog circuit design, with talks exploring innovations in parasitic capacitance prediction for AMS circuits. The third presentation shifts focus to sequential circuit design, introducing novel methods for behavior-centric optimization through machine learning techniques. Next, a physical design recipe recommender is presented, showcasing how AI can enhance the design process. Finally, the session concludes with two talks on simulation, examining the integration of machine learning with SPICE models and the use of neural compact modeling for accelerating design-technology co-development.