Conference proceeding
TensorFI: A Configurable Fault Injector for TensorFlow Applications
2018 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW), pp.313-320
10/2018
DOI: 10.1109/ISSREW.2018.00024
Abstract
Machine Learning (ML) applications have emerged as the killer applications for next generation hardware and software platforms, and there is a lot of interest in software frameworks to build such applications. TensorFlow is a high-level dataflow framework for building ML applications and has become the most popular one in the recent past. ML applications are also being increasingly used in safety-critical systems such as self-driving cars and home robotics. Therefore, there is a compelling need to evaluate the resilience of ML applications built using frameworks such as TensorFlow. In this paper, we build a high-level fault injection framework for TensorFlow called TensorFI for evaluating the resilience of ML applications. TensorFI is flexible, easy to use, and portable. It also allows ML application programmers to explore the effects of different parameters and algorithms on error resilience.
Details
- Title: Subtitle
- TensorFI: A Configurable Fault Injector for TensorFlow Applications
- Creators
- Guanpeng Li - University of British ColumbiaKarthik Pattabiraman - University of British ColumbiaNathan DeBardeleben - Los Alamos National Laboratory
- Resource Type
- Conference proceeding
- Publication Details
- 2018 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW), pp.313-320
- DOI
- 10.1109/ISSREW.2018.00024
- Publisher
- IEEE
- Language
- English
- Date published
- 10/2018
- Academic Unit
- Computer Science
- Record Identifier
- 9984259431202771
Metrics
18 Record Views