Out-of-Distribution Detection in High-Dimensional Data Using Mahalanobis Distance - Critical Analysis | SpringerLink
Skip to main content

Out-of-Distribution Detection in High-Dimensional Data Using Mahalanobis Distance - Critical Analysis

  • Conference paper
  • First Online:
Computational Science – ICCS 2022 (ICCS 2022)

Abstract

Convolutional neural networks used in real-world recognition must be able to detect inputs that are Out-of-Distribution (OoD) with respect to the known or training data. A popular, simple method is to detect OoD inputs using confidence scores based on the Mahalanobis distance from known data. However, this procedure involves estimating the multivariate normal (MVN) density of high dimensional data using the insufficient number of observations (e.g., the dimensionality of features at the last two layers in the ResNet-101 model are 2048 and 1024, with ca. 1000–5000 examples per class for density estimation). In this work, we analyze the instability of parametric estimates of MVN density in high dimensionality and analyze the impact of this on the performance of Mahalanobis distance-based OoD detection. We show that this effect makes Mahalanobis distance-based methods ineffective for near OoD data. We show that the minimum distance from known data beyond which outliers are detectable depends on the dimensionality and number of training samples and decreases with the growing size of the training dataset. We also analyzed the performance of modifications of the Mahalanobis distance method used to minimize density fitting errors, such as using a common covariance matrix for all classes or diagonal covariance matrices. On OoD benchmarks (on CIFAR-10, CIFAR-100, SVHN, and Noise datasets), using representations from the DenseNet or ResNet models, we show that none of these methods should be considered universally superior.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save

Springer+ Basic
¥17,985 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
JPY 3498
Price includes VAT (Japan)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
JPY 12583
Price includes VAT (Japan)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
JPY 15729
Price includes VAT (Japan)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    https://www.cs.toronto.edu/~kriz/cifar.html.

References

  1. Bendale, A., Boult, T.: Towards Open World Recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1893–1902 (2015)

    Google Scholar 

  2. Breunig, M.M., Kriegel, H.P., Ng, R.T., Sander, J.: Lof: identifying density-based local outliers. SIGMOD Rec. 29(2), 93–104 (2000). https://doi.org/10.1145/335191.335388

    Article  Google Scholar 

  3. Chakraborty, A., Alam, M., Dey, V., Chattopadhyay, A., Mukhopadhyay, D.: A survey on adversarial attacks and defences. CAAI Trans. Intell. Technol. 6(1), 25–45 (2021). https://doi.org/10.1049/cit2.12028

  4. Eykholt, K., et al.: Robust physical-world attacks on deep learning visual classification. In: 2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, 18–22 June 2018, pp. 1625–1634. IEEE Computer Society (2018). https://doi.org/10.1109/CVPR.2018.00175

  5. Feng, D., Rosenbaum, L., Dietmayer, K.: Towards safe autonomous driving: capture uncertainty in the deep neural network for lidar 3d vehicle detection. In: 2018 21st International Conference on Intelligent Transportation Systems (ITSC), pp. 3266–3273. IEEE (2018)

    Google Scholar 

  6. Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, New York (2009). https://doi.org/10.1007/978-0-387-21606-5

  7. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)

    Google Scholar 

  8. Hendrycks, D., Mazeika, M., Dietterich, T.: Deep anomaly detection with outlier exposure. Proceedings of the International Conference on Learning Representations (2019)

    Google Scholar 

  9. Hendrycks, D., Zhao, K., Basart, S., Steinhardt, J., Song, D.: Natural adversarial examples. arXiv preprint arXiv:1907.07174 (2019)

  10. Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4700–4708 (2017)

    Google Scholar 

  11. Lee, K., Lee, K., Lee, H., Shin, J.: A simple unified framework for detecting out-of-distribution samples and adversarial attacks. In: Proceedings of the 32nd International Conference on Neural Information Processing Systems, pp. 7167–7177. NIPS 2018, Curran Associates Inc., Red Hook, NY (2018)

    Google Scholar 

  12. Lin, M., Chen, Q., Yan, S.: Network in network. arXiv preprint arXiv:1312.4400 (2013)

  13. McAllister, R., et al.: Concrete problems for autonomous vehicle safety: Advantages of bayesian deep learning. In: Proceedings of the 26th International Joint Conference on Artificial Intelligence, pp. 4745–4753. IJCAI 2017, AAAI Press (2017)

    Google Scholar 

  14. Nguyen, A., Yosinski, J., Clune, J.: Deep neural networks are easily fooled: high confidence predictions for unrecognizable images. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 427–436 (2015)

    Google Scholar 

  15. Pham, H., Dai, Z., Xie, Q., Luong, M.T., Le, Q.V.: Meta pseudo labels. In: IEEE Conference on Computer Vision and Pattern Recognition (2021). https://arxiv.org/abs/2003.10580

  16. Ren, J., Fort, S., Liu, J., Roy, A.G., Padhy, S., Lakshminarayanan, B.: A simple fix to mahalanobis distance for improving near-ood detection. arXiv preprint arXiv:2106.09022 (2021)

  17. Rousseeuw, P.J.: Least median of squares regression. J. Am. Stat. Assoc. 79(388), 871–880 (1984)

    Article  MathSciNet  Google Scholar 

  18. Sehwag, V., Chiang, M., Mittal, P.: Ssd: a unified framework for self-supervised outlier detection. In: International Conference on Learning Representations (2021). https://openreview.net/forum?id=v5gjXpmR8J

  19. Sharif, M., Bhagavatula, S., Bauer, L., Reiter, M.K.: Accessorize to a crime: real and stealthy attacks on state-of-the-art face recognition. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, pp. 1528–1540 (2016)

    Google Scholar 

  20. Zhou, Z., Firestone, C.: Humans can decipher adversarial images. Nature Commun. 10(1), 1–9 (2019)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Henryk Maciejewski .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Maciejewski, H., Walkowiak, T., Szyc, K. (2022). Out-of-Distribution Detection in High-Dimensional Data Using Mahalanobis Distance - Critical Analysis. In: Groen, D., de Mulatier, C., Paszynski, M., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M.A. (eds) Computational Science – ICCS 2022. ICCS 2022. Lecture Notes in Computer Science, vol 13350. Springer, Cham. https://doi.org/10.1007/978-3-031-08751-6_19

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-08751-6_19

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-08750-9

  • Online ISBN: 978-3-031-08751-6

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics