[HTML][HTML] Resilience of deep learning applications: A systematic literature review of analysis and hardening techniques

C Bolchini, L Cassano, A Miele - Computer Science Review, 2024 - Elsevier
Abstract Machine Learning (ML) is currently being exploited in numerous applications, being
one of the most effective Artificial Intelligence (AI) technologies used in diverse fields, such …

Resilience of Deep Learning applications: a systematic literature review of analysis and hardening techniques

C Bolchini, L Cassano, A Miele - arXiv preprint arXiv:2309.16733, 2023 - arxiv.org
Machine Learning (ML) is currently being exploited in numerous applications being one of
the most effective Artificial Intelligence (AI) technologies, used in diverse fields, such as …

Baywatch: Leveraging bayesian neural networks for hardware fault tolerance and monitoring

J Hoefer, M Stammler, F Kreß, T Hotfilter… - … on Defect and Fault …, 2024 - ieeexplore.ieee.org
As Deep Neural Networks are increasingly being utilized in safety-critical domains,
assessing the uncertainty of the models during inference will be a crucial component in …

Leveraging mixed-precision CNN inference for increased robustness and energy efficiency

T Hotfilter, J Hoefer, P Merz, F Kreß… - 2023 IEEE 36th …, 2023 - ieeexplore.ieee.org
Convolutional Neural Networks (CNNs) show tremendous performance in many Computer
Vision (CV) tasks like image segmentation crucial to autonomous driving. However, they are …

[HTML][HTML] ALPRI-FI: A Framework for Early Assessment of Hardware Fault Resiliency of DNN Accelerators

K Mahmoud, N Nicolici - Electronics, 2024 - mdpi.com
Understanding how faulty hardware affects machine learning models is important to both
safety-critical systems and the cloud infrastructure. Since most machine learning models …

Single-Event Upset Analysis of a Systolic Array based Deep Neural Network Accelerator

N Jonckers, T Vinck, G Dekkers, P Karsmakers… - arXiv preprint arXiv …, 2024 - arxiv.org
Deep Neural Network (DNN) accelerators are extensively used to improve the
computational efficiency of DNNs, but are prone to faults through Single-Event Upsets …

ZuSE-KI-Mobil AI Chip Design Platform: An Overview

S Mojumder, S Friedrich, E Matúš… - 2024 IEEE Nordic …, 2024 - ieeexplore.ieee.org
The ZuSE-KI-Mobil (ZuKIMo) project, a nationally funded initiative, focuses on creating an
advanced ecosystem optimized for AI-driven applications in automotive, drone, and …

ZuSE-KI-mobil: Platform for Energy Efficient AI-Processors in Mobile Applications

HJ Voegel, J Becker, J Benndorf… - MikroSystemTechnik …, 2023 - ieeexplore.ieee.org
ZuSE KI-mobil is a development platform for applications that use methods of" Artificial
Intelligence" to solve complex applications. The concept includes the development of a SOC …