The National Science Foundation Division of Computing and Communication Foundations awarded the University of Southern California $411,682 on October 1, 2025, under the Computer and Information Science and Engineering program to develop memory-efficient algorithms and hardware co-design for spiking neural networks deployed on power-constrained edge computing devices.
The project aims to integrate spiking neural networks with modern integrated circuits to reduce power consumption in artificial intelligence applications including object detection, autonomous driving, and image classification. The research team will devise novel algorithms and hardware designs with prototype chips to accelerate spiking neural network performance in low-power, memory-efficient systems. The resulting neuromorphic chips are intended to enable practical deployment in drones, autonomous robots, portable medical devices, and wearable smart assistants. The project also encompasses workforce development and training for high school, undergraduate, and graduate students in neuromorphic computing, with emphasis on diversity and inclusivity in AI and semiconductor research fields.
Performance occurs at the University of Southern California in Los Angeles, California, with a period of performance from October 1, 2025, through September 30, 2027.