Method for Hybrid Precision Convolutional Neural Network Representation

Autor:	Al-Hami, Mo'taz, Pietron, Marcin, Kumar, Rishi, Casas, Raul A., Hijazi, Samer L., Rowen, Chris
Rok vydání:	2018
Předmět:	Computer Science - Neural and Evolutionary Computing
Druh dokumentu:	Working Paper
Popis:	This invention addresses fixed-point representations of convolutional neural networks (CNN) in integrated circuits. When quantizing a CNN for a practical implementation there is a trade-off between the precision used for operations between coefficients and data and the accuracy of the system. A homogenous representation may not be sufficient to achieve the best level of performance at a reasonable cost in implementation complexity or power consumption. Parsimonious ways of representing data and coefficients are needed to improve power efficiency and throughput while maintaining accuracy of a CNN. Comment: Cadence Design Systems
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1807.09760 Zobrazit plný text záznamu View this record from Arxiv