Presentation

DescriptionChip design and fabrication is at a crucial stage with increasing complexity, customization, and constraints. Innovative tools and ideas are critical for the development of next generation EDA tools. Machine Learning is one such tool that both contributes to better EDA tools and simultaneously benefits from the increased computing power that these EDA tools can make available. This session includes two talks that discuss how the use of ML enhances physical design and fabrication. The third talk of the session focuses on how ML algorithms can be accelerated on heterogenous compute systems.