A Long Short-Term Memory for AI Applications in Spike-based Neuromorphic Hardware

Spike-based neuromorphic hardware holds promise for more energy-efficient implementations of deep neural networks (DNNs) than standard hardware such as GPUs. But this requires us to understand how DNNs can be emulated in an event-based sparse firing regime, as otherwise the energy advantage is lost....

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Nature machine intelligence 2022-05, Vol.4 (5), p.467-479
Hauptverfasser:	Rao, Arjun, Plank, Philipp, Wild, Andreas, Maass, Wolfgang
Format:	Artikel
Sprache:	eng
Schlagworte:	631/114/1305 631/114/2405 631/378/116/1925 Artificial neural networks Energy consumption Energy efficiency Engineering Hardware Machine learning Memory Neural networks Neuromorphic computing Neurons Time series
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!