Structural Test Coverage Criteria for Deep Neural Networks
- Submitting institution
-
The University of Liverpool
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 12200
- Type
- D - Journal article
- DOI
-
10.1145/3358233
- Title of journal
- ACM Transactions on Embedded Computing Systems
- Article number
- 94
- First page
- 1
- Volume
- 18
- Issue
- 5S
- ISSN
- 1539-9087
- Open access status
- Not compliant
- Month of publication
- October
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
5
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The preliminary arXiv version of this paper has the title "Testing Deep Neural Networks". The paper proposes a way to adapt the MC/DC coverage metric (a test coverage metric promoted by NASA and required by industry standards such as ISO26262 and RTCA DO-187B/C) for deep neural networks. Since its publication, Huang and co-authors have developed this proposal further, for example in "Concolic Testing for Deep Neural Networks" (ASE 2018), not REF returned. Major research contributions that have been directly influenced by this work include Odena and Goodfellow "TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing" (ICML 2019).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -