New Grant: Award Supports Design Automation for Real-Time Computing

April 14, 2022 –

Principal investigator: Mohammad Al Faruque, associate professor of electrical engineering and computer science, and co-principal investigator Marco Levorato, associate professor of computer science

Award: $499,998 over three years

Funding agency: National Science Foundation Division of Computing and Communication Foundations

Project: A Design Automation Methodology for Flexible Real-Time Computing Based on Split and Early Exit Neural Models

Al Faruque is developing highly adaptable deep-learning (DL) frameworks for real-time applications, such as autonomous vehicles and mobile health. The proposed design-automation methodology will bridge runtime system optimization with advanced DL model architectures through leveraging the techniques of split computing and early exit computation. The new frameworks will build DL models specifically designed to adapt real-time data analysis to time-varying characteristics of the system (for example, available energy, computing power, channel capacity, computing task, etc.) and the information stream. The team will use tools such as deep reinforcement learning and neural architecture search. The project will include an educational and outreach plan for undergraduate and graduate students and proposes a suite of university initiatives that will enhance diversity.

More information: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2140154&HistoricalAwards=false

– Tonya Becerra