Quantized Neural Network with Linearly Approximated Functions on Zynq FPGA

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F24%3A00375910" target="_blank" >RIV/68407700:21240/24:00375910 - isvavai.cz</a>

Result on the web

<a href="https://doi.org/10.1109/MECO62516.2024.10577851" target="_blank" >https://doi.org/10.1109/MECO62516.2024.10577851</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/MECO62516.2024.10577851" target="_blank" >10.1109/MECO62516.2024.10577851</a>

Result language

angličtina

Original language name

Quantized Neural Network with Linearly Approximated Functions on Zynq FPGA

Original language description

This paper is focused on neural network implementation on FPGA. Linearly approximated functions combined with quantization are used to efficiently implement neural networks in hardware. Famous benchmarks were used for learning, evaluation, and hardware testing. Approximation-aware and quantization-aware learning were used to obtain weights for neurons in hardware. We implemented a neural network with an 8-bit architecture in VHDL and synthesized it to Zynq FPGA in Vivado. The resulting design running at 100MHz clock frequency was carefully tested against hardware-accurate models written in Wolfram Mathematica and C++. We present a decrease in FPGA resources and chip utilization compared to 16-bit architecture implementation.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2024

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

2024 13th Mediterranean Conference on Embedded Computing (MECO)

ISBN

979-8-3503-8756-8

ISSN

2377-5475

e-ISSN

2637-9511

Number of pages

4

Pages from-to

98-101

Publisher name

Institute of Electrical and Electronic Engineers

Place of publication

Piscataway

Event location

Budva

Event date

Jun 11, 2024

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

001268606200050