A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks Paper Title: A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks Link to Paper: https://arxiv.org/abs/2305.05750 Date: 9. May 2023 Paper Type: Neural Network Reliability, Fault Simulation, Fault Injection, Literature Review Short Abstract: This paper reviews methods for assessing the reliability of Deep Neural Networks (DNNs), especially in safety-critical applications where hardware errors can have serious consequences. It presents a systematic literature review categorizing reliability assessment approaches into Fault Injection (FI), Analytical, and Hybrid methods. The study explains the strengths, weaknesses, platforms, and evaluation metrics of these methods, highlighting that while FI is the most common approach, Analytical and Hybrid methods are more lightweight and still accurate, making them promising directions for future DNN reliability research.

MRFI: An Open Source Multi-Resolution Fault Injection Framework for Neural Network Processing Paper Title: MRFI: An Open Source Multi-Resolution Fault Injection Framework for Neural Network Processing Link to Paper: https://arxiv.org/abs/2306.11758 Date: 12. Dec. 2023 Paper Type: Neural Network Reliability, Fault Simulation, Fault Injection Short Abstract: MRFI is a highly configurable, multi-resolution fault injection tool for deep neural networks, designed to address the limitations of existing fault injection solutions. 1. Introduction Deep Neural Networks (DNNs) are increasingly deployed in safety-critical applications (e.g., autonomous driving, avionics) and large-scale systems (e.g., LLMs), where hardware faults—such as process variations, defects, noise, or bit flips—can cause erroneous inferences with potentially severe consequences. Therefore, thorough reliability evaluation through fault injection before deployment is essential.

Runtime Fault Localization in Deep Neural Network Accelerators Paper Title: Runtime Fault Localization in Deep Neural Network Accelerators Link to Paper: https://dl.acm.org/doi/10.1145/3770920 Date: 11. Nov 2025 Paper Type: Systolic arrays, deep neural network (DNN) accelerator, fault detection, fault localization, Checksums Short Abstract: Systolic arrays are widely used for accelerating deep neural networks (DNNs) due to their high parallelism and data reuse efficiency. However, hardware faults in their numerous processing elements (PEs) can propagate errors and significantly degrade inference accuracy. The paper addresses the open challenge of fault localization in systolic arrays. It proposes a lightweight fault tolerance framework that performs both run-time fault detection and localization during normal operation. The approach uses functional data to generate checksums on-the-fly, eliminating the need for dedicated test patterns or system downtime. Reuslts: 100% fault detection and localization, 2% Area overhead.