The DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index.
Why ReLU Networks Yield High-Confidence Predictions Far Away From the Training Data and How to Mitigate the Problem
Open accessAbstract
Classifiers used in the wild, in particular for safety-critical systems, should not only have good generalization properties but also should know when they don't know, in particular make low confidence predictions far away from the training data. We show that ReLU type neural networks which yield a piecewise linear classifier function fail in this regard as they produce almost always high confidence predictions far away from the training data. For bounded domains like images we propose a new robust optimization technique similar to adversarial training which enforces low confidence predictions far away from the training data. We show that this technique is surprisingly effective in reducing the confidence of predictions far away from the training data while maintaining high confidence predictions and test error on the original classification task compared to standard training.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
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
Intriguing properties of neural networks
Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1312.6199 · Cited by 8,143
Towards Deep Learning Models Resistant to Adversarial Attacks
Madry, A., Makelov, A., Schmidt, L., Tsipras, D., & Vladu, A. (2017) · arXiv · doi:10.48550/arxiv.1706.06083 · Cited by 6,602
Dropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471
Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
Xiao, H., Rasul, K., & Vollgraf, R. (2017) · arXiv · doi:10.48550/arxiv.1708.07747 · Cited by 6,108
DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
Moosavi-Dezfooli, S.-M., Fawzi, A., & Frossard, P. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.282 · Cited by 5,317
Cited by
Shortcut learning in deep neural networks
Geirhos, R., Jacobsen, J.-H., Michaelis, C., Zemel, R., Brendel, W., Bethge, M., & Wichmann, F. A. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-020-00257-z · Cited by 2,224
A survey of uncertainty in deep neural networks
Gawlikowski, J., Tassi, C. R. N., Ali, M., Lee, J., Humt, M., Feng, J., Kruspe, A., Triebel, R., Jung, P., Roscher, R., et al. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10562-9 · Cited by 1,406
Activation functions in deep learning: a comprehensive survey and benchmark
Dubey, S. R., Singh, S. K., & Chaudhuri, B. B. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.06.111 · Cited by 1,259
Dynamic Neural Networks: A Survey
Han, Y., Huang, G., Song, S., Yang, L., Wang, H., & Wang, Y. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3117837 · Cited by 848
Generalized Out-of-Distribution Detection: A Survey
Yang, J., Zhou, K., Li, Y., & Liu, Z. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02117-4 · Cited by 884
Energy-based Out-of-distribution Detection
Liu, W., Wang, X., Owens, J. D., & Li, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.03759 · Cited by 412
Uncertainty Estimation Using a Single Deep Deterministic Neural Network
van Amersfoort, J., Smith, L., Teh, Y. W., & Gal, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.02037 · Cited by 263
Simple and principled uncertainty estimation with deterministic deep learning via distance awareness
Liu, J. Z., Lin, Z., Padhy, S., Tran, D., Bedrax-Weiss, T., & Lakshminarayanan, B. (2020) · arXiv · doi:10.48550/arxiv.2006.10108 · Cited by 237
Salehi, M., Mirzaei, H., Hendrycks, D., Li, Y., Rohban, M. H., & Sabokrou, M. (2021) · arXiv · doi:10.48550/arxiv.2110.14051 · Cited by 118
Explainable Artificial Intelligence (XAI) on TimeSeries Data: A Survey
Rojat, T., Puget, R., Filliat, D., Del Ser, J., Gelin, R., & Díaz-Rodríguez, N. (2021) · arXiv · doi:10.48550/arxiv.2104.00950 · Cited by 116
Mixing up contrastive learning: Self-supervised representation learning for time series
Wickstrøm, K., Kampffmeyer, M., Mikalsen, K. Ø., & Jenssen, R. (2022) · Pattern Recognition Letters · doi:10.1016/j.patrec.2022.02.007 · Cited by 113
Beyond generalization: a theory of robustness in machine learning
Freiesleben, T., & Grote, T. (2023) · Synthese · doi:10.1007/s11229-023-04334-9 · Cited by 106
Artificial intelligence in safety-critical systems: a systematic review
Wang, Y., & Chung, S. H. (2022) · Industrial Management & Data Systems · doi:10.1108/imds-07-2021-0419 · Cited by 96
Mix-n-match: Ensemble and compositional methods for uncertainty calibration in deep learning
Zhang, J., Kailkhura, B., & Han, Y.-J. (2020) · Unpublished · doi:10.13140/rg.2.2.11390.13125 · Cited by 86
AI system engineering key challenges and lessons learned
Fischer, L., Ehrlinger, L., Geist, V., Ramler, R., Sobiezky, F., Zellinger, W., Brunner, D., Kumar, M., & Moser, B. (2020) · Machine Learning and Knowledge Extraction · doi:10.3390/make3010004 · Cited by 71
“However, functions like softmax can result in extrapolations with unjustified high confidence for points far from the training data, hence providing a false sense of safety [18].”
BackgroundTaxonomy of Machine Learning Safety: A Survey and Primer
Mohseni, S., Wang, H., Xiao, C., Yu, Z., Wang, Z., & Yadawa, J. (2023) · ACM Computing Surveys · doi:10.1145/3551385 · Cited by 53
Quantifying uncertainty in deep learning approaches to radio galaxy classification
Mohan, D., Scaife, A. M. M., Porter, F., Walmsley, M., & Bowles, M. (2022) · Monthly Notices of the Royal Astronomical Society · doi:10.1093/mnras/stac223 · Cited by 16
Interpretable AI/ML for High-stakes Tasks with Human-in-the-loop: Critical Review and Future Trends
Kovalerchuk, B. (2024) · doi:10.21203/rs.3.rs-3989807/v1 · Cited by 9
Bringing Machine Learning Classifiers Into Critical Cyber-Physical Systems: A Matter of Design
Sayin, B., Zoppi, T., Marchini, N., Khokhar, F. A., & Passerini, A. (2025) · IEEE Access · doi:10.1109/access.2025.3568501 · Cited by 7
“Overconfidence is common among both humans and machines, especially in modern DNNs [25].”
Metadata
- Authors
- Hein, M., Andriushchenko, M., & Bitterwolf, J.
- Venue
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2019-06
- Pages
- 41-50
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2905485203
- Cited by
- 428
- References
- 36
- FWCI
- 21.44
- Type
- conference paper
- Language
- en
- Open access
- green
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.