Prediction of Inference Energy on CNN Accelerators Supporting Approximate Circuits

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F23%3APU148180" target="_blank" >RIV/00216305:26230/23:PU148180 - isvavai.cz</a>

Result on the web

<a href="https://ieeexplore.ieee.org/document/10139724" target="_blank" >https://ieeexplore.ieee.org/document/10139724</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/DDECS57882.2023.10139724" target="_blank" >10.1109/DDECS57882.2023.10139724</a>

Result language

angličtina

Original language name

Prediction of Inference Energy on CNN Accelerators Supporting Approximate Circuits

Original language description

Design methodologies developed for optimizing hardware implementations of convolutional neural networks (CNN) or searching for new hardware-aware neural architectures rely on the fast and reliable estimation of key hardware parameters, such as the energy needed for one inference. Utilizing approximate circuits in hardware accelerators of CNNs faces the designers with new problems during their simulation - commonly used tools (TimeLoop, Accelergy, Maestro) do not support approximate arithmetic operations. This work addresses the fast and efficient prediction of consumed energy in hardware accelerators of CNNs that utilize approximate circuits such as approximate multipliers. First, we extend the state-of-the-art software frameworks TimeLoop and Accelergy to predict the inference energy when exact multipliers are replaced with various approximate implementations. The energies obtained using the modified tools are then considered the ground truth (reference) values. Then, we propose and evaluate, using two accelerators (Eyeriss and Simba) and two types of networks (CNNs generated by EvoApproxNAS and standard ResNet CNNs), two predictors of inference energy. We conclude that a simple predictor based on summing the energies needed for all multiplications highly correlates with the reference values if the CNN's architecture is fixed. For complex CNNs with variable architectures typically generated by neural architecture search algorithms, a more sophisticated predictor based on a machine learning model has to be employed. The proposed predictors are 420-533× faster than reference solutions.

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

<a href="/en/project/GA22-02067S" target="_blank" >GA22-02067S: AppNeCo: Approximate Neurocomputing</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

Article name in the collection

2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems

ISBN

979-8-3503-3277-3

ISSN

—

e-ISSN

—

Number of pages

6

Pages from-to

45-50

Publisher name

Institute of Electrical and Electronics Engineers

Place of publication

Talinn

Event location

Tallinn

Event date

May 3, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

001012062000008