Approximate computing techniques offer a promising solution to reduce the hardware complexity and power consumption imposed when embedding machine learning algorithms. The reduction comes at the cost of some performance degradation. This paper presents an approximate machine learning classifier for touch modality recognition. Experimental results demonstrate that the use of software level approximation techniques reduce the execution time and memory usage up to 38% and 55% respectively, at the cost of accuracy loss less than 10% for the target application. © 2019 IEEE.
Data Oriented Approximate K-Nearest Neighbor Classifier for Touch Modality Recognition
Younes H.;Ibrahim A.;Valle M.
2019-01-01
Abstract
Approximate computing techniques offer a promising solution to reduce the hardware complexity and power consumption imposed when embedding machine learning algorithms. The reduction comes at the cost of some performance degradation. This paper presents an approximate machine learning classifier for touch modality recognition. Experimental results demonstrate that the use of software level approximation techniques reduce the execution time and memory usage up to 38% and 55% respectively, at the cost of accuracy loss less than 10% for the target application. © 2019 IEEE.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
08787753.pdf
accesso chiuso
Descrizione: file in post print
Tipologia:
Documento in versione editoriale
Dimensione
913.99 kB
Formato
Adobe PDF
|
913.99 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.