Acceleration Techniques for Automated Design of Approximate Convolutional Neural Networks

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F23%3APU146796" target="_blank" >RIV/00216305:26230/23:PU146796 - isvavai.cz</a>

Result on the web

<a href="https://ieeexplore.ieee.org/document/10011413" target="_blank" >https://ieeexplore.ieee.org/document/10011413</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/JETCAS.2023.3235204" target="_blank" >10.1109/JETCAS.2023.3235204</a>

Result language

angličtina

Original language name

Acceleration Techniques for Automated Design of Approximate Convolutional Neural Networks

Original language description

The main issue connected with using approximate components such as approximate multipliers in deep convolutional neural networks (CNN) during the design process is the necessity to emulate them due to the lack of native support for approximate operations in modern CPUs and GPUs, which is computationally expensive. To accelerate the emulation of approximate operations of CNNs on GPUs, we propose TFApprox4IL, a software library supporting both symmetric as well as asymmetric quantization modes, approximate 8xN bit multipliers emulated using lookup tables, a new type of approximate layer known as approximate depthwise convolution, and quantization-aware training. The TFApprox4IL performance is extensively evaluated in the simulation of approximate implementations of MobileNetV2 and ResNet networks on Nvidia Pascal and Tesla GPU architectures. Furthermore, TFApprox4IL is also evaluated in neural architecture search (NAS) algorithms to automatically design CNN architectures that directly employ approximate multipliers. On two different NAS methods, EvoApproxNAS and Google Model Search (GMS), we show how approximate multipliers can effectively be incorporated into the CNN design process. To estimate the energy consumption of the approximate CNNs, AxMultAT tool based on Timeloop and Accelergy is introduced. Contrasted to the highly optimized GPU-based CNN simulation implemented using exact arithmetic operations available within TensorFlow, the average overhead of the inference and training, introduced by TFApprox4IL, is 13.6× and 8.0× , respectively, considering ResNet50V2 and MobileNetV2 CNNs on ImageNet and CIFAR-10 data sets. This overhead was reduced by one order of magnitude with respect to previous methods.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

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

Project

<a href="/en/project/GA21-13001S" target="_blank" >GA21-13001S: Automated design of hardware accelerators for resource-aware machine learning</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2023

Confidentiality

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

Name of the periodical

IEEE Journal on Emerging and Selected Topics in Circuits and Systems

ISSN

2156-3357

e-ISSN

—

Volume of the periodical

13

Issue of the periodical within the volume

1

Country of publishing house

US - UNITED STATES

Number of pages

13

Pages from-to

212-224

UT code for WoS article

000965262200001

EID of the result in the Scopus database

2-s2.0-85147308000