Training progressively Binarizing Deep Networks using FPGAs

Conference Publication ResearchOnline@JCU
Lammie, Corey;Xiang, Wei;Rahimi Azghadi, Mostafa
Abstract

While hardware implementations of inference routines for Binarized Neural Networks (BNNs) are plentiful, current realizations of efficient BNN hardware training accelerators, suitable for Internet of Things (IoT) edge devices, leave much to be desired. Conventional BNN hardware training accelerators perform forward and backward propagations with parameters adopting binary representations, and optimization using parameters adopting floating or fixed-point real-valued representations-requiring two distinct sets of network parameters. In this paper, we propose a hardware-friendly training method that, contrary to conventional methods, progressively binarizes a singular set of fixed-point network parameters, yielding notable reductions in power and resource utilizations. We use the Intel FPGA SDK for OpenCL development environment to train our progressively binarizing DNNs on an OpenVINO FPGA. We benchmark our training approach on both GPUs and FPGAs using CIFAR-10 and compare it to conventional BNNs.

Journal

N/A

Publication Name

2020 IEEE International Symposium on Circuits and Systems (ISCAS)

Volume

N/A

ISBN/ISSN

978-1-7281-3320-1

Edition

N/A

Issue

N/A

Pages Count

5

Location

Seville, Spain

Publisher

Institute of Electrical and Electronics Engineers

Publisher Url

N/A

Publisher Location

Piscataway, NJ, USA

Publish Date

N/A

Url

N/A

Date

N/A

EISSN

N/A

DOI

10.1109/ISCAS45731.2020.9181099