Is neuron coverage a meaningful measure for testing deep neural networks?
Open accessThe DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
Recent effort to test deep learning systems has produced an intuitive and compelling test criterion called neuron coverage (NC), which resembles the notion of traditional code coverage. NC measures the proportion of neurons activated in a neural network and it is implicitly assumed that increasing NC improves the quality of a test suite. In an attempt to automatically generate a test suite that increases NC, we design a novel diversity promoting regularizer that can be plugged into existing adversarial attack algorithms. We then assess whether such attempts to increase NC could generate a test suite that (1) detects adversarial attacks successfully, (2) produces natural inputs, and (3) is unbiased to particular class predictions. Contrary to expectation, our extensive evaluation finds that increasing NC actually makes it harder to generate an effective test suite: higher neuron coverage leads to fewer defects detected, less natural inputs, and more biased prediction preferences. Our results invoke skepticism that increasing neuron coverage may not be a meaningful objective for generating tests for deep neural networks and call for a new test generation technique that considers defect detection, naturalness, and output impartiality in tandem.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
On Information and Sufficiency
Kullback, S., & Leibler, R. A. (1951) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177729694 · Cited by 20,193
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612
Hinton, G., Deng, L., Yu, D., Dahl, G., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V., Nguyen, P., Sainath, T., et al. (2012) · IEEE Signal Processing Magazine · doi:10.1109/msp.2012.2205597 · Cited by 10,260 · cited 1×
“DNNs have achievedmany breakthroughs in the field of artificial intelligence, such as speech recognition [18], image processing [28], statistical machine translation [3], and game playing [55].”
Towards Evaluating the Robustness of Neural Networks
Carlini, N., & Wagner, D. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.49 · Cited by 8,859
Improved Techniques for Training GANs
Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., & Chen, X. (2016) · arXiv · doi:10.48550/arxiv.1606.03498 · Cited by 6,965
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
The Limitations of Deep Learning in Adversarial Settings
Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z. B., & Swami, A. (2016) · 2016 IEEE European Symposium on Security and Privacy (EuroS&P) · doi:10.1109/eurosp.2016.36 · Cited by 3,993
The correlation coefficient: its values range between +1/-1, or do they?
Ratner, B. (2009) · Journal of Targeting, Measurement and Analysis for Marketing · doi:10.1057/jt.2009.5 · Cited by 1,612
Olah, C., Mordvintsev, A., & Schubert, L. (2017) · Distill · doi:10.23915/distill.00007 · Cited by 880
Zhang, M., Zhang, Y., Zhang, L., Liu, C., & Khurshid, S. (2018) · Proceedings of the 33rd ACM/IEEE International Conference on Automated Software Engineering · doi:10.1145/3238147.3238187 · Cited by 611
Fairness testing: testing software for discrimination
Galhotra, S., Brun, Y., & Meliou, A. (2017) · Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering · doi:10.1145/3106237.3106277 · Cited by 405 · cited 1×
“[12], automatically detects causal discrimination between input-output pairs for user-specified attributes.”
Background
Cited by
Giray, G. (2021) · Journal of Systems and Software · doi:10.1016/j.jss.2021.111031 · Cited by 187
Perez-Cerrolaza, J., Abella, J., Borg, M., Donzella, C., Cerquides, J., Cazorla, F. J., Englund, C., Tauber, M., Nikolakopoulos, G., & Flores, J. L. (2024) · ACM Computing Surveys · doi:10.1145/3626314 · Cited by 185
Free lunch for testing: fuzzing deep-learning libraries from open source
Wei, A., Deng, Y., Yang, C., & Zhang, L. (2022) · Proceedings of the 44th International Conference on Software Engineering · doi:10.1145/3510003.3510041 · Cited by 106
“metrics and the robustness of models is still unclear [25, 36, 69].”
BackgroundMethodFuzzing deep-learning libraries via automated relational API inference
Deng, Y., Yang, C., Wei, A., & Zhang, L. (2022) · Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering · doi:10.1145/3540250.3549085 · Cited by 86
“To date, most prior work on DL testing focused on testing/verifying DL models, with an emphasis on adversarial attacks [14, 19, 27, 42, 47, 51], metrics for model testing [31, 34, 41, 53, 77], application-”
Metadata
- Authors
- Harel-Canada, F., Wang, L., Gulzar, M. A., Gu, Q., & Kim, M.
- Venue
- Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering
- Publisher
- ACM
- Published
- 2020-11-08
- Pages
- 851-862
- OpenAlex
- W2994987245
- Cited by
- 145
- References
- 14
- FWCI
- 14.27
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.