PULP-NN: accelerating quantized neural networks on parallel ultra-low-power RISC-V processors

We present PULP-NN, an optimized computing library for a parallel ultra-low-power tightly coupled cluster of RISC-V processors. The key innovation in PULP-NN is a set of kernels for quantized neural network inference, targeting byte and sub-byte data types, down to INT-1, tuned for the recent trend...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Philosophical transactions of the Royal Society of London. Series A: Mathematical, physical, and engineering sciences physical, and engineering sciences, 2020-02, Vol.378 (2164), p.20190155-20190155
Hauptverfasser:	Garofalo, Angelo, Rusci, Manuele, Conti, Francesco, Rossi, Davide, Benini, Luca
Format:	Artikel
Sprache:	eng
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!