Characterizing Deep Neural Networks Neutrons-Induced Error Model

Fernandes dos Santos, Fernando
Kritikakou, Angeliki
Sentieys, Olivier
Rech, Paolo

Publication date

July 2022

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Abstract

International audienceWe characterize the fault models for Deep Neural Networks (DNNs) in GPUs exposed to neutron. We observe tolerable and critical errors, and show that ECC is not effective in reducing critical errors

Extracted data

We use cookies to provide a better user experience.

Data Protection

Characterizing Deep Neural Networks Neutrons-Induced Error Model

Abstract

Extracted data

Characterizing Deep Neural Networks Neutrons-Induced Error Model

Abstract

Extracted data

Related items

Related items