Conference proceeding
TensorFI: A Flexible Fault Injection Framework for TensorFlow Applications
2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE), Vol.2020-, pp.426-435
10/2020
DOI: 10.1109/ISSRE5003.2020.00047
Abstract
As machine learning (ML) has seen increasing adoption in safety-critical domains (e.g., autonomous vehicles), the reliability of ML systems has also grown in importance. While prior studies have proposed techniques to enable efficient error-resilience (e.g., selective instruction duplication), a fundamental requirement for realizing these techniques is a detailed understanding of the application's resilience. In this work, we present TensorFI, a high-level fault injection (FI) framework for TensorFlow-based applications. TensorFI is able to inject both hardware and software faults in general TensorFlow programs. TensorFI is a configurable FI tool that is flexible, easy to use, and portable. It can be integrated into existing TensorFlow programs to assess their resilience for different fault types (e.g., faults in particular operators). We use TensorFI to evaluate the resilience of 12 ML programs, including DNNs used in the autonomous vehicle domain. The results give us insights into why some of the models are more resilient. We also present two case studies to demonstrate the usefulness of the tool. TensorFI is publicly available at https://github.com/DependableSystemsLab/TensorFI.
Details
- Title: Subtitle
- TensorFI: A Flexible Fault Injection Framework for TensorFlow Applications
- Creators
- Zitao Chen - University of British ColumbiaNiranjhana Narayanan - University of British ColumbiaBo Fang - University of British ColumbiaGuanpeng Li - University of IowaKarthik Pattabiraman - University of British ColumbiaNathan DeBardeleben - Los Alamos National Laboratory
- Resource Type
- Conference proceeding
- Publication Details
- 2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE), Vol.2020-, pp.426-435
- Publisher
- IEEE
- DOI
- 10.1109/ISSRE5003.2020.00047
- ISSN
- 1071-9458
- eISSN
- 2332-6549
- Grant note
- Natural Sciences and Engineering Research Council of Canada (10.13039/501100000038)
- Language
- English
- Date published
- 10/2020
- Academic Unit
- Computer Science
- Record Identifier
- 9984259497702771
Metrics
11 Record Views