EUVIMEDEUVIMED entdecken
Crossref

Adaptive Soft Error Protection for Neural Network Processing

Xinghua Xue, Cheng Liu, Feng Min, Hui Li, Kai Zhang, Yinhe Han

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · 2026-11 · 45 (11) · 6490-6500

DOI: 10.1109/tcad.2026.3667089 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2026-11. Verfasst von Xinghua Xue und 5 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Previous research on selective protection for neural network components typically exploits only static vulnerability differences. Although these methods improve upon classical modular redundancy, they still incur substantial overhead for neural network workloads that are both memory-intensive and compute-intensive. In this work, we observe that neural network vulnerability is also input-dependent and varies dynamically at runtime. With this observation, we propose an adaptive, vulnerability-aware fault tolerance framework. At its core, a lightweight graph neural network (GNN) model dynamically predicts soft error vulnerabilities across inputs and neural network components, enabling real-time adaptation of fault tolerance policies. This design offers a complementary and more efficient protection scheme compared to traditional approaches. Experimental results demonstrate that the GNN predictor achieves over 95% accuracy in identifying critical inputs and components. Moreover, our adaptive scheme reduces computational overhead by an average of 42.12% while preserving model accuracy, significantly outperforming static selective protection methods.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Xue, X., Liu, C., Min, F., Li, H., Zhang, K., & Han, Y. (2026). Adaptive Soft Error Protection for Neural Network Processing. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 45(11), 6490-6500. https://doi.org/10.1109/tcad.2026.3667089

Vancouver

Xue X, Liu C, Min F, Li H, Zhang K, Han Y. Adaptive Soft Error Protection for Neural Network Processing. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. 2026;45(11):6490-6500. doi:10.1109/tcad.2026.3667089

BibTeX
@article{Xue2026,
  author = {Xue, Xinghua and Liu, Cheng and Min, Feng and Li, Hui and Zhang, Kai and Han, Yinhe},
  title = {Adaptive Soft Error Protection for Neural Network Processing},
  journal = {IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems},
  year = {2026},
  volume = {45},
  number = {11},
  pages = {6490--6500},
  doi = {10.1109/tcad.2026.3667089}
}
Kostenlos starten →