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
| Autor: | Dewei Wang, Mingoo Seok, Sang Joon Kim, Pavan Kumar Chundi, Sung Justin Kim, Joonsung Kang, Minhao Yang, Seungchul Jung, Joao P. Cerqueira |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Spiking neural network
FOS: Computer and information sciences Computer Science - Machine Learning Power gating Artificial neural network Computer science Real-time computing Computer Science - Neural and Evolutionary Computing Power (physics) Machine Learning (cs.LG) Low-power electronics Keyword spotting Spike (software development) Power semiconductor device Neural and Evolutionary Computing (cs.NE) |
| Zdroj: | A-SSCC |
| DOI: | 10.48550/arxiv.2006.12314 |
| Popis: | 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 [1]. 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|