• Home
  • Search
  • Comparing Application-Level Hardening Techniques for Neural Networks on GPUs
  • https://doi.org/10.3390/electronics14051042Copy DOI Icon

Comparing Application-Level Hardening Techniques for Neural Networks on GPUs

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Neural networks (NNs) are essential in advancing modern safety-critical systems. Lightweight NN architectures are deployed on resource-constrained devices using hardware accelerators like Graphics Processing Units (GPUs) for fast responses. However, the latest semiconductor technologies may be affected by physical faults that can jeopardize the NN computations, making fault mitigation crucial for safety-critical domains. The recent studies propose software-based Hardening Techniques (HTs) to address these faults. However, the proposed fault countermeasures are evaluated through different hardware-agnostic error models neglecting the effort required for their implementation and different test benches. Comparing application-level HTs across different studies is challenging, leaving it unclear (i) their effectiveness against hardware-aware error models on any NN and (ii) which HTs provide the best trade-off between reliability enhancement and implementation cost. In this study, application-level HTs are evaluated homogeneously and independently by performing a study on the feasibility of implementation and a reliability assessment under two hardware-aware error models: (i) weight single bit-flips and (ii) neuron bit error rate. Our results indicate that not all HTs suit every NN architecture, and their effectiveness varies depending on the evaluated error model. Techniques based on the range restriction of activation function consistently outperform others, achieving up to 58.23% greater mitigation effectiveness while keeping the introduced overhead at inference time low while requiring a contained effort in their implementation.

Similar Papers
  • Research Article
  • Citations9

Evaluation of the computational efficacy in GPU-accelerated simulations of spiking neurons

  • Feb 01, 2018
  • Computing
  • Kazuhisa Fujita +2
  • Book Chapter
  • Citations19

Introduction to hardware accelerator systems for artificial intelligence and machine learning

  • Sep 14, 2020
  • Neha Gupta
  • Book Chapter
  • Citations1

RTL Verification and FPGA Implementation of Generalized Neural Networks: A High-Level Synthesis Approach

  • Jan 01, 2022
  • Satyashil D Nagarale +1
  • Research Article
  • Citations1

A parallel approach to accelerate neural network hyperparameter selection for energy forecasting

  • Jun 01, 2025
  • Expert Systems with Applications
  • D Criado-Ramón +2
  • Book Chapter
  • Citations7

GPU Acceleration of Sparse Neural Networks

  • Jan 01, 2021
  • Aavaas Gajurel +4
  • Dissertation

Neural Visual Synthesis for Graphics, 4D Gaussians and Videos

  • Mar 13, 2025
  • Zhan Li
  • Research Article
  • Citations2

Обнаружение очагов возгорания на технологических объектах с использованием сверточной нейронной сети

  • Oct 25, 2023
  • Pozharovzryvobezopasnost/Fire and Explosion Safety
  • А А Евсиков +1
  • Conference Article
  • Citations9

High-performance computing for automatic target recognition in synthetic aperture radar imagery

  • May 01, 2017
  • Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE
  • Uttam Majumder +5
  • Research Article
  • Citations19

A Virtual Multi-Channel GPU Fair Scheduling Method for Virtual Machines

  • Feb 01, 2019
  • IEEE Transactions on Parallel and Distributed Systems
  • Huailiang Tan +4
  • PDF
  • Research Article
  • Citations76

NnAudio: An on-the-Fly GPU Audio to Spectrogram Conversion Toolbox Using 1D Convolutional Neural Networks

  • Jan 01, 2020
  • IEEE Access
  • Kin Wai Cheuk +3
  • PDF
  • Research Article
  • Citations24

Integrated photonic FFT for photonic tensor operations towards efficient and high-speed neural networks

  • Jun 26, 2020
  • Nanophotonics
  • Moustafa Ahmed +4
  • Conference Article
  • Citations4

Fast Biological Sequence Comparison on Hybrid Platforms

  • Sep 01, 2014
  • Safia Kedad-Sidhoum +4
  • Research Article
  • Citations1

WE‐C‐108‐05: A Hardware Accelerator Based Fast Monte Carlo Code for Radiation Dosimetry: Software Design and Preliminary Results

  • Jun 01, 2013
  • Medical Physics
  • X Du +4
  • Conference Article
  • Citations75

Large scale recurrent neural network on GPU

  • Jul 01, 2014
  • Boxun Li +7
  • Conference Article
  • Citations136

Neural Network Implementation Using CUDA and OpenMP

  • Jan 01, 2008
  • Honghoon Jang +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.