The reliability of Neural Networks has gained significant attention, prompting efforts to develop SW-based hardening techniques for safety-critical scenarios. However, evaluating hardening techniques using application-level fault injection (FI) strategies, which are commonly hardware-agnostic, may yield misleading results. This study for the first time compares two FI approaches (at the application level (APP) and instruction level (ISA)) to evaluate deep neural network SW hardening strategies. Results show that injecting permanent faults at ISA (a more detailed abstraction level than APP) changes completely the ranking of SW hardening techniques, in terms of both reliability and accuracy. These results highlight the relevance of using an adequate analysis abstraction for evaluating such techniques.
Evaluating Different Fault Injection Abstractions on the Assessment of DNN SW Hardening Strategies / Esposito, Giuseppe; Guerrero Balaguera, Juan David; Rodriguez Condia, Josie Esteban; Sonza Reorda, Matteo. - ELETTRONICO. - (In corso di stampa), pp. 1-6. (Intervento presentato al convegno 33rd IEEE Asian Test Symposium (ATS 2024) tenutosi a Ahmedabad, Gujarat (IND) nel 17th -20th December 2024).
Evaluating Different Fault Injection Abstractions on the Assessment of DNN SW Hardening Strategies
Esposito, Giuseppe;Guerrero Balaguera, Juan David;Rodriguez Condia, Josie Esteban;Sonza Reorda ,Matteo
In corso di stampa
Abstract
The reliability of Neural Networks has gained significant attention, prompting efforts to develop SW-based hardening techniques for safety-critical scenarios. However, evaluating hardening techniques using application-level fault injection (FI) strategies, which are commonly hardware-agnostic, may yield misleading results. This study for the first time compares two FI approaches (at the application level (APP) and instruction level (ISA)) to evaluate deep neural network SW hardening strategies. Results show that injecting permanent faults at ISA (a more detailed abstraction level than APP) changes completely the ranking of SW hardening techniques, in terms of both reliability and accuracy. These results highlight the relevance of using an adequate analysis abstraction for evaluating such techniques.File | Dimensione | Formato | |
---|---|---|---|
_ATS2024_Eval_Hard_App_NvbitPERfi-2.pdf
accesso aperto
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Pubblico - Tutti i diritti riservati
Dimensione
535.69 kB
Formato
Adobe PDF
|
535.69 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2996511