Presentation
EPHA: An Energy-efficient Parallel Hybrid Architecture for AGI via Neuromorphic Compensated Ferrimagnets
DescriptionArtificial neural networks (ANNs) and spiking neural networks (SNNs) are two general approaches to achieve artificial general intelligence (AGI). This paper presents an energy-efficient, scalable, and non-von neumann architecture (EPHA) for ANNs and SNNs. Our study combines device-, circuit-, architecture-, and algorithm-level innovations to achieve a parallel architecture with ultra-low power consumption. We use the compensated ferrimagnet to act as both synapses and neurons to store weights and perform dot-product operations. In the ANN mode, the EPHA is 10× more energy-efficient than the NEBULA. In the SNN mode, our design is 3 orders of magnitude more energy-efficient than the TrueNorth.
Event Type
Work-in-Progress Poster
TimeWednesday, July 12th6:00pm - 7:00pm PDT
LocationLevel 2 Lobby
AI
Autonomous Systems
Cloud
Design
EDA
Embedded Systems
RISC-V
Security