Multi-Objective Application-Driven Approximate Design Method

Autor: Salvatore Barone, Marcello Traiola, Mario Barbareschi, Alberto Bosio

Publikováno v: IEEE Access, Vol 9, Pp 86975-86993 (2021)

Approximate Computing (AxC) paradigm aims at designing computing systems that can satisfy the rising performance demands and improve the energy efficiency. AxC exploits the gap between the level of accuracy required by the users, and the actual preci

Externí odkaz: https://doaj.org/article/408cd901548747898f5a807e26ee445d

A Survey on Deep Learning Resilience Assessment Methodologies

Autor: Annachiara Ruospo, Ernesto Sanchez, Lucas Matana Luza, Luigi Dilillo, Marcello Traiola, Alberto Bosio

Publikováno v: Computer
Computer, 2023, 56, pp.57-66. ⟨10.1109/MC.2022.3217841⟩

International audience; Deep Learning (DL) applications are gaining increasing interest in the industry and academia for their outstanding computational capabilities. Indeed, they have found successful applications in various areas and domains such a

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fe77c451d906e07fc429717fa472b3c3
https://doi.org/10.1109/mc.2022.3217841

Selective Hardening of Critical Neurons in Deep Neural Networks

Autor: Annachiara Ruospo, Gabriele Gavarini, Ilaria Bragaglia, Marcello Traiola, Alberto Bosio, Ernesto Sanchez

Publikováno v: DDECS 2022-25th International Symposium on Design and Diagnostics of Electronic Circuits and Systems
DDECS 2022-25th International Symposium on Design and Diagnostics of Electronic Circuits and Systems, Apr 2022, Prague, Czech Republic. pp.136-141, ⟨10.1109/DDECS54261.2022.9770168⟩

International audience; In the literature, it is argued that Deep Neural Networks (DNNs) possess a certain degree of robustness mainly for two reasons: their distributed and parallel architecture, and their redundancy introduced due to over provision

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e7f7cebf3e60ad9e8183680acd20d5c3
https://hal.inria.fr/hal-03888005/document

Input-Aware Approximate Computing

Autor: Ali Piri, Sepide Saeedi, Mario Barbareschi, Bastien Deveautour, Stefano Di Carlo, Ian O'Connor, Alessandro Savino, Marcello Traiola, Alberto Bosio

Publikováno v: AQTR 2022-IEEE International Conference on Automation, Quality and Testing, Robotics
AQTR 2022-IEEE International Conference on Automation, Quality and Testing, Robotics, May 2022, Cluj-Napoca, Romania. pp.1-6, ⟨10.1109/AQTR55203.2022.9801944⟩
2022 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR)

International audience; In the last decade, Approximate Computing (AxC) has been extensively employed to improve the energy efficiency of computing systems, at different abstraction levels. The main AxC goal is reducing the energy budget used to exec

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::df9b54ecb33ad77d8b59f34caf41af7e
https://hdl.handle.net/11588/915830

Pros and Cons of Fault Injection Approaches for the Reliability Assessment of Deep Neural Networks

Autor: Annachiara Ruospo, Lucas Matana Luza, Alberto Bosio, Marcello Traiola, Luigi Dilillo, Ernesto Sanchez

Publikováno v: LATS 2021-IEEE 22nd Latin American Test Symposium
LATS 2021-IEEE 22nd Latin American Test Symposium, Oct 2021, Punta del Este, Uruguay. pp.1-5, ⟨10.1109/LATS53581.2021.9651807⟩
IEEE Latin-American Test Symposium (LATS 2021)
IEEE Latin-American Test Symposium (LATS 2021), Oct 2021, Punta del Este, Uruguay

International audience; In the last years, the adoption of Artificial Neural Networks (ANNs) in safety-critical applications has required an in-depth study of their reliability. For this reason, the research community has shown a growing interest in

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0972e236925cd6c74d6f97eab89ab7f6
https://hal-lirmm.ccsd.cnrs.fr/lirmm-03435567