FORTALESA: Fault-tolerant reconfigurable systolic array for DNN inference
Authors:
- Natalia CherezovaNCNatalia Cherezova
- Artur JutmanAJArtur Jutman
- Maksim JenihhinMJMaksim Jenihhin
Open Access
- 30 October 2025
- journal article
- Published by Elsevier in Microprocessors and Microsystems
- Vol. 119
Abstract
No abstract availableAll Related Versions
Funding Information
- Horizon Europe
- Estonian Research Council (TK202)
- Estonian Research Council (PUT PRG1467)
This publication has 44 references indexed in Scilit:
- RePAIR: Reconfigurable Platform for AI Resilience Within RISC-V Ecosystem2025
- Error Checking for Sparse Systolic Tensor Arrays2024
- A Low-Cost Strategic Monitoring Approach for Scalable and Interpretable Error Detection in Deep Neural Networks2023
- Enhancing Fault Resilience of QNNs by Selective Neuron Splitting2023
- Saca-FI: A microarchitecture-level fault injection framework for reliability analysis of systolic array based CNN acceleratorFuture Generation Computer Systems, 2023
- A Comprehensive Analysis of Transient Errors on Systolic Arrays2023
- Gemmini: Enabling Systematic Deep-Learning Architecture Evaluation via Full-Stack Integration2021
- SHIELDeNN: Online Accelerated Framework for Fault-Tolerant Deep Neural Network Architectures2020
- Resilient Neural Network Training for Accelerators with Computing Errors2019
- Statistical fault injection: Quantified error and confidence2009