Always-On, Sub-300-nW, Event-Driven Spiking Neural Network based on Spike-Driven Clock-Generation and Clock- and Power-Gating for an Ultra-Low-Power Intelligent Device

Always-on artificial intelligent (AI) functions such as keyword spotting (KWS) and visual wake-up tend to dominate total power consumption in ultra-low power devices. A key observation is that the signals to an always-on function are sparse in time, which a spiking neural network (SNN) classifier ca...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-06
Main Authors: Wang, Dewei, Chundi, Pavan Kumar, Sung Justin Kim, Yang, Minhao, Joao Pedro Cerqueira, Kang, Joonsung, Jung, Seungchul, Kim, Sangjoon, Mingoo Seok
Format: Article
Language:English
Subjects:
Artificial intelligence
Classifiers
Electronic devices
Neural networks
Power consumption
Power management
Spikes
Spiking
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Always-on artificial intelligent (AI) functions such as keyword spotting (KWS) and visual wake-up tend to dominate total power consumption in ultra-low power devices. A key observation is that the signals to an always-on function are sparse in time, which a spiking neural network (SNN) classifier can leverage for power savings, because the switching activity and power consumption of SNNs tend to scale with spike rate. Toward this goal, we present a novel SNN classifier architecture for always-on functions, demonstrating sub-300nW power consumption at the competitive inference accuracy for a KWS and other always-on classification workloads.
ISSN:2331-8422