FKeras: A Fault Tolerance Library for Keras

Olivia Weng; Andres Meza; Javier M. Duarte; Nhan Tran; Ryan Kastner

doi:10.1364/3D.2023.JTu4A.40

Optica Imaging Congress (3D, COSI, DH, FLatOptics, IS, pcAOP)
Technical Digest Series (Optica Publishing Group, 2023),
paper JTu4A.40
•https://doi.org/10.1364/3D.2023.JTu4A.40

FKeras: A Fault Tolerance Library for Keras

Olivia Weng, Andres Meza, Javier M. Duarte, Nhan Tran, and Ryan Kastner

3D Image Acquisition and Display: Technology, Perception and Applications 2023

Boston, Massachusetts United States
14–17 August 2023
ISBN: 978-1-957171-28-9

From the session
Joint Poster Session I (JTu4A)

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
O. Weng, A. Meza, J. M. Duarte, N. Tran, and R. Kastner, "FKeras: A Fault Tolerance Library for Keras," in Optica Imaging Congress (3D, COSI, DH, FLatOptics, IS, pcAOP), Technical Digest Series (Optica Publishing Group, 2023), paper JTu4A.40.

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

We present FKeras, an open-source tool that uses Hessian information to quickly find which parameters in a neural network are sensitive to radiation faults, reducing the usual 200% resource overhead needed to protect them.

PDF Article

More Like This

Fault-Tolerant Computing on POEM

Dau-Tsuong Lu, Ting-Ting Y. Lin, Fouad E. Kiamilev, Sadik C. Esener, and Sing H. Lee
ME12 Optical Computing (IP) 1991

Highly fault-tolerant quantum computing using both discrete and continuous variables of light

Hyunseok Jeong, Seok-Hyung Lee, Srikrishna Omkar, and Yong Siah Teo
QTu4A.6 Quantum 2.0 (QUANTUM) 2023

Operator Quantum Fault Tolerance

Gerald Gilbert, Michael Hamrick, Yaakov S. Weinstein, Vaneet Aggarwal, and A. Robert Calderbank
JMB46 International Conference on Quantum Information (QIM) 2008

Previous Article Next Article

Poster Presentation

Media 1: PDF (933 KB)