TY - GEN
T1 - PyTorchFI
T2 - 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN-W 2020
AU - Mahmoud, Abdulrahman
AU - Aggarwal, Neeraj
AU - Nobbe, Alex
AU - Vicarte, Jose Rodrigo Sanchez
AU - Adve, Sarita V.
AU - Fletcher, Christopher W.
AU - Frosio, Iuri
AU - Hari, Siva Kumar Sastry
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/6
Y1 - 2020/6
N2 - PyTorchFI is a runtime perturbation tool for deep neural networks (DNNs), implemented for the popular PyTorch deep learning platform. PyTorchFI enables users to perform perturbations on weights or neurons of DNNs at runtime. It is designed with the programmer in mind, providing a simple and easy-to-use API, requiring as little as three lines of code for use. It also provides an extensible interface, enabling researchers to choose from various perturbation models (or design their own custom models), which allows for the study of hardware error (or general perturbation) propagation to the software layer of the DNN output. Additionally, PyTorchFI is extremely versatile: we demonstrate how it can be applied to five different use cases for dependability and reliability research, including resiliency analysis of classification networks, resiliency analysis of object detection networks, analysis of models robust to adversarial attacks, training resilient models, and for DNN interpertability. This paper discusses the technical underpinnings and design decisions of PyTorchFI which make it an easy-to-use, extensible, fast, and versatile research tool. PyTorchFI is open-sourced and available for download via pip or github at: https://github.com/pytorchfi
AB - PyTorchFI is a runtime perturbation tool for deep neural networks (DNNs), implemented for the popular PyTorch deep learning platform. PyTorchFI enables users to perform perturbations on weights or neurons of DNNs at runtime. It is designed with the programmer in mind, providing a simple and easy-to-use API, requiring as little as three lines of code for use. It also provides an extensible interface, enabling researchers to choose from various perturbation models (or design their own custom models), which allows for the study of hardware error (or general perturbation) propagation to the software layer of the DNN output. Additionally, PyTorchFI is extremely versatile: we demonstrate how it can be applied to five different use cases for dependability and reliability research, including resiliency analysis of classification networks, resiliency analysis of object detection networks, analysis of models robust to adversarial attacks, training resilient models, and for DNN interpertability. This paper discusses the technical underpinnings and design decisions of PyTorchFI which make it an easy-to-use, extensible, fast, and versatile research tool. PyTorchFI is open-sourced and available for download via pip or github at: https://github.com/pytorchfi
UR - http://www.scopus.com/inward/record.url?scp=85092635000&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092635000&partnerID=8YFLogxK
U2 - 10.1109/DSN-W50199.2020.00014
DO - 10.1109/DSN-W50199.2020.00014
M3 - Conference contribution
AN - SCOPUS:85092635000
T3 - Proceedings - 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN-W 2020
SP - 25
EP - 31
BT - Proceedings - 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN-W 2020
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 29 June 2020 through 2 July 2020
ER -