Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks

Eyeriss is an accelerator for state-of-the-art deep convolutional neural networks (CNNs). It optimizes for the energy efficiency of the entire system, including the accelerator chip and off-chip DRAM, for various CNN shapes by reconfiguring the architecture. CNNs are widely used in modern AI systems...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE journal of solid-state circuits 2017-01, Vol.52 (1), p.127-138
Hauptverfasser:	Yu-Hsin Chen, Krishna, Tushar, Emer, Joel S., Sze, Vivienne
Format:	Artikel
Sprache:	eng
Schlagworte:	Clocks Computer architecture Convolutional neural networks (CNNs) dataflow processing deep learning energy-efficient accelerators Hardware Neural networks Random access memory Reconfiguration Shape spatial architecture Throughput
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!