Hardware-based Spiking Neural Network
Lead Research Organisation:
Brunel University London
Department Name: Electronic and Electrical Engineering
Abstract
Inspired by biological brains, spiking neural networks (SNN) are considered the third
generation of artificial neural networks (ANNs) (Maass, 1997). SNNs use recurrent computing graphs and sparse binary signals to process information, rather than real numbers used by ANNs. SNNs demonstrate several advantages over ANNs such as biological plausibility, energy efficiency, improved latency, increased speed and computational power (Bing, 2018; Lobov, 2020; Lee, 2016). Another major advantage of SNNs is that they can have a self-learning ability, which will enable them to better cope in dynamic environments. For example, walking on cobblestones which then become wet.
Responding to dynamic environments is something that classical control and ANNs often struggle to do.
generation of artificial neural networks (ANNs) (Maass, 1997). SNNs use recurrent computing graphs and sparse binary signals to process information, rather than real numbers used by ANNs. SNNs demonstrate several advantages over ANNs such as biological plausibility, energy efficiency, improved latency, increased speed and computational power (Bing, 2018; Lobov, 2020; Lee, 2016). Another major advantage of SNNs is that they can have a self-learning ability, which will enable them to better cope in dynamic environments. For example, walking on cobblestones which then become wet.
Responding to dynamic environments is something that classical control and ANNs often struggle to do.
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/T518116/1 | 01/10/2020 | 30/09/2025 | |||
| 2691294 | Studentship | EP/T518116/1 | 01/01/2022 | 31/12/2024 | TIM FERNANDEZ-HART |