2
IRUS Total
Downloads
  Altmetric

CascadeC(NN): pushing the performance limits of quantisation in convolutional neural networks

File Description SizeFormat 
1807.05053v1.pdfAccepted version4.29 MBAdobe PDFView/Open
Title: CascadeC(NN): pushing the performance limits of quantisation in convolutional neural networks
Authors: Kouris, A
Venieris, SI
Bouganis, C-S
Item Type: Conference Paper
Abstract: This work presents CascadeCNN, an automated toolflow that pushes the quantisation limits of any given CNN model, aiming to perform high-throughput inference. A two-stage architecture tailored for any given CNN-FPGA pair is generated, consisting of a low-and high-precision unit in a cascade. A confidence evaluation unit is employed to identify misclassified cases from the excessively low-precision unit and forward them to the high-precision unit for re-processing. Experiments demonstrate that the proposed toolflow can achieve a performance boost up to 55% for VGG-16 and 48% for AlexNet over the baseline design for the same resource budget and accuracy, without the need of retraining the model or accessing the training data.
Issue Date: 6-Dec-2018
Date of Acceptance: 27-Aug-2018
URI: http://hdl.handle.net/10044/1/69927
DOI: https://dx.doi.org/10.1109/FPL.2018.00034
ISSN: 1946-1488
Publisher: IEEE
Start Page: 155
End Page: 162
Journal / Book Title: 2018 28th International Conference on Field Programmable Logic and Applications (FPL)
Copyright Statement: © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Conference Name: 28th International Conference on Field Programmable Logic and Applications (FPL)
Keywords: Science & Technology
Technology
Computer Science, Hardware & Architecture
Computer Science, Software Engineering
Computer Science
cs.CV
cs.LG
Publication Status: Published
Start Date: 2018-08-26
Finish Date: 2018-12-06
Conference Place: Dublin, IRELAND
Online Publication Date: 2018-12-06
Appears in Collections:Faculty of Engineering
Electrical and Electronic Engineering