Low-Cost Online Convolution Checksum Checker

Filippas, Dionysios; Margomenos, Nikolaos; Mitianoudis, Nikolaos; Nicopoulos, Chrysostomos; Dimitrakopoulos, Giorgos

doi:10.1109/tvlsi.2021.3119511

Cited by 13 publications

(11 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Algorithm-Based Fault Tolerance (ABFT): Another line of research leverages algorithmic perspective to make DNNs reliable. These methods, known as ABFT [21], [22], [23], [24], typically compute checksums for input data, which are then stored alongside the original data. Then, both the original and redundant computations are conducted to verify outputs.…”

Section: Related Workmentioning

confidence: 99%

Analysis and Enhancement of Resilience for LSTM Accelerators Using Residue-Based CEDs

Nosrati,

Navabi

2024

IEEE Access

View full text Add to dashboard Cite

As Long Short-Term Memory (LSTM) accelerators are increasingly being employed in safetycritical applications with high-reliability demands, protecting them against errors becomes imperative. Traditional protection techniques for LSTMs are either costly, conflict with strict area and power constraints for neural network accelerators, or introduce performance overhead that is untenable for real-time and latency-critical accelerators. In this paper, we propose residue-based Concurrent Error Detection (CED) schemes to detect transient faults and alleviate their impact during LSTM computations. CED units are employed in a coarse-grain or fine-grain fashion, depending on the granularity of the components being protected from the whole LSTM computations to individual processing elements. In pursuit of a more costeffective strategy, we use fine-grain CEDs in a selective manner based on the resilience characteristics of LSTM. The selections are made spatially for LSTM synaptic weights or temporally based on LSTM time steps. The experimental results show that the proposed residue-based CEDs (i) can achieve nearly complete fault coverage even under extremely large bit error rates, (ii) significantly decrease misprediction rates compared to the unprotected LSTM, and (iii) incur low overhead without compromising performance. Our method is compared with modular redundancy techniques such as DMR and TMR (Double and Triple Modular Redundancy).

show abstract

Section: Related Workmentioning

confidence: 99%

Analysis and Enhancement of Resilience for LSTM Accelerators Using Residue-Based CEDs

Nosrati,

Navabi

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To reduce the overhead, a lightweight ABFT, namely Con-vGuard has been proposed by state-of-the-art research [83], which only predicts the convolution checksum on the border pixels of the input feature map. ConvGuard computes the checksum of the output pixels and compares it to its predicted checksum value in parallel to the operation of the convolution engine.…”

Section: B Algorithm-level Resiliencementioning

confidence: 99%

“…FIGURE 10. Generalized ConvGuard architecture that supports arbitrary convolutions (modified from[83]). …”

mentioning

confidence: 99%

Dependable DNN Accelerator for Safety-Critical Systems: A Review on the Aging Perspective

Moghaddasi,

Gorgin,

Lee

2023

IEEE Access

View full text Add to dashboard Cite

Nowadays, artificial intelligence (AI) and deep learning (DL) progressively adapt to various spheres of our lives. These disciplines contain safety-critical applications such as autonomous driving with a high risk of human injury in the case of malfunction, requiring a high promise of dependability. Even the dependability becomes more crucial as shrinking CMOS technology feature size worsens the resilience concerns due to factors like aging. This paper addresses the overarching dependability issue of advanced deep neural networks (DNN) accelerators from the aging perspective. Especially, a comprehensive survey and taxonomy of techniques used to evaluate and mitigate aging effects are introduced. We cover different aging effects like permanent faults, timing errors, and lifetime issues. We review research by the layer-wise approach and categorize several resilience classes to bring out major features. The concluding part of this review highlights the questions answered and several future research directions. This study is expected to benefit researchers in different areas of DNN deployment, especially the dependability of this emergent paradigm.

show abstract

“…Huang and Abraham introduced Algorithm Based Fault Tolerance (ABFT) for matrix algebra operation [17]. ABFT has been demonstrated to be robust and attractive with overhead ratio of O(1/N ), with N being the size of the matrices.…”

Section: B Algorithmic Schemesmentioning

confidence: 99%