THIS GRANT IS A CONTINUATION OF N000141510031 EFFICIENT NEURO-COMPUTING USING SPIN-TORQUE DEVICES
Abstract
We propose a research program to investigate spin-based neuromorphic computing from the device levelto new computing model/architecture. At the device level, we will develop magneto-metallic spindevices that mimic certain biological component functionality in a highly energy efficient manner. Theproposed devices are based on spin-torque experiments that have demonstrated the feasibility ofstructures with perpendicular magnetic anisotropy, domain wall motion enhanced by spin hall effect,lateral spin valves. We will also explore systems that integrate such spin-devices with CMOS andSilicon memristors to realize complete neural network/ other biological system functionality. A devicecircuit-architecture co-simulation-framework (benchmarked against available experiments) will bedeveloped to evaluate functionality and performance at various levels of abstraction. Our preliminaryevaluations indicates the possibility of more than 100X improvements in energy consumption forapplications such as image recognition, compared to state-of-the-art CMOS. The goals of the proposedresearch are to (i) propose devices and establish the benefits of spin based devices in the context ofneuromorphic/biological computing (for which CMOS implementations are energy-inefficient), (ii)synergistically explore spin devices, circuits and architecture to augment capabilities of CMOS, and (iii)incorporate the findings in education and outreach programs.If successful, spin devices enable new computing models at ultra-low energy consumption, mobiledevices to run richer, more complex applications such as image/target recognition, video processing.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2016
- Source ID
- N000141612518
Entities
People
- Kaushik Roy
Organizations
- Office of Naval Research
- United States Navy
- University of Virginia