Advertisement

ConvNet and Dempster-Shafer Theory for Object Recognition

  • Zheng Tong
  • Philippe Xu
  • Thierry DenœuxEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11940)

Abstract

We propose a novel classifier based on convolutional neural network (ConvNet) and Dempster-Shafer theory for object recognition allowing for ambiguous pattern rejection, called the ConvNet-BF classifier. In this classifier, a ConvNet with nonlinear convolutional layers and a global pooling layer extracts high-dimensional features from input data. The features are then imported into a belief function classifier, in which they are converted into mass functions and aggregated by Dempster’s rule. Evidence-theoretic rules are finally used for pattern classification and rejection based on the aggregated mass functions. We propose an end-to-end learning strategy for adjusting the parameters in the ConvNet and the belief function classifier simultaneously and determining the rejection loss for evidence-theoretic rules. Experiments with the CIFAR-10, CIFAR-100, and MNIST datasets show that hybridizing belief function classifiers with ConvNets makes it possible to reduce error rates by rejecting patterns that would otherwise be misclassified.

Keywords

Pattern recognition Belief function Convolutional neural network Supervised learning Evidence theory 

References

  1. 1.
    Bengio, Y.: Learning deep architectures for AI. Found. Trends® Mach. Learn. 2(1), 1–127 (2009)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Bi, Y.: The impact of diversity on the accuracy of evidential classifier ensembles. Int. J. Approximate Reasoning 53(4), 584–607 (2012)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Dempster, A.P.: Upper and lower probabilities induced by a multivalued mapping. In: Yager, R.R., Liu, L. (eds.) Classic Works of the Dempster-Shafer Theory of Belief Functions. STUDFUZZ, vol. 219, pp. 57–72. Springer, Heidelberg (2008).  https://doi.org/10.1007/978-3-540-44792-4_3CrossRefGoogle Scholar
  4. 4.
    Denœux, T.: A k-nearest neighbor classification rule based on Dempster-Shafer theory. IEEE Trans. Syst. Man Cybern. 25(5), 804–813 (1995)CrossRefGoogle Scholar
  5. 5.
    Denœux, T.: Analysis of evidence-theoretic decision rules for pattern classification. Pattern Recogn. 30(7), 1095–1107 (1997)CrossRefGoogle Scholar
  6. 6.
    Denœux, T.: A neural network classifier based on Dempster-Shafer theory. IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 30(2), 131–150 (2000)CrossRefGoogle Scholar
  7. 7.
    Denœux, T.: Logistic regression, neural networks and Dempster-Shafer theory: a new perspective. Knowl.-Based Syst. 176, 54–67 (2019)CrossRefGoogle Scholar
  8. 8.
    Denœux, T., Dubois, D., Prade, H.: Representations of uncertainty in artificial intelligence: beyond probability and possibility. In: Marquis, P., Papini, O., Prade, H. (eds.) A Guided Tour of Artificial Intelligence Research, Chap. 4. Springer (2019)Google Scholar
  9. 9.
    Denœux, T., Kanjanatarakul, O., Sriboonchitta, S.: A new evidential K-nearest neighbor rule based on contextual discounting with partially supervised learning. Int. J. Approximate Reasoning 113, 287–302 (2019)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Gomez, A.N., Zhang, I., Swersky, K., Gal, Y., Hinton, G.E.: Targeted dropout. In: CDNNRIA Workshop at the 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), Montréal (2018)Google Scholar
  11. 11.
    Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.R.: Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 (2012)
  12. 12.
    Kim, Y.: Convolutional neural networks for sentence classification. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, Doha, pp. 1746–1751 (2014)Google Scholar
  13. 13.
    Krizhevsky, A., Hinton, G.: Learning multiple layers of features from tiny images. University of Toronto, Technical report (2009)Google Scholar
  14. 14.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017)CrossRefGoogle Scholar
  15. 15.
    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436 (2015)CrossRefGoogle Scholar
  16. 16.
    LeCun, Y., Bottou, L., Bengio, Y., Haffner, P., et al.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)CrossRefGoogle Scholar
  17. 17.
    Leng, B., Liu, Y., Yu, K., Zhang, X., Xiong, Z.: 3D object understanding with 3D convolutional neural networks. Inf. Sci. 366, 188–201 (2016)MathSciNetCrossRefGoogle Scholar
  18. 18.
    Lin, M., Chen, Q., Yan, S.: Network in network. In: International Conference on Learning Representations (ICLR 2014), Banff, pp. 1–10 (2014)Google Scholar
  19. 19.
    Liu, Z., Pan, Q., Dezert, J., Han, J.W., He, Y.: Classifier fusion with contextual reliability evaluation. IEEE Trans. Cybern. 48(5), 1605–1618 (2018)CrossRefGoogle Scholar
  20. 20.
    Minary, P., Pichon, F., Mercier, D., Lefevre, E., Droit, B.: Face pixel detection using evidential calibration and fusion. Int. J. Approximate Reasoning 91, 202–215 (2017)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Sakaguchi, K., Post, M., Van Durme, B.: Efficient elicitation of annotations for human evaluation of machine translation. In: Proceedings of the Ninth Workshop on Statistical Machine Translation, Baltimore, pp. 1–11 (2014)Google Scholar
  22. 22.
    Salakhutdinov, R., Hinton, G.: Deep Boltzmann machines. In: Artificial Intelligence and Statistics, Florida, pp. 448–455 (2009)Google Scholar
  23. 23.
    Salakhutdinov, R., Tenenbaum, J.B., Torralba, A.: Learning with hierarchical-deep models. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1958–1971 (2012)CrossRefGoogle Scholar
  24. 24.
    Shafer, G.: A Mathematical Theory of Evidence. Princeton University Press, Princeton (1976)zbMATHGoogle Scholar
  25. 25.
    Tong, Z., Gao, J., Zhang, H.: Recognition, location, measurement, and 3D reconstruction of concealed cracks using convolutional neural networks. Constr. Build. Mater. 146, 775–787 (2017)CrossRefGoogle Scholar
  26. 26.
    Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th International Conference on Machine Learning, pp. 1096–1103, New York (2008)Google Scholar
  27. 27.
    Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., Manzagol, P.A.: Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J. Mach. Learn. Res. 11(Dec), 3371–3408 (2010)MathSciNetzbMATHGoogle Scholar
  28. 28.
    Xu, P., Davoine, F., Zha, H., Denœux, T.: Evidential calibration of binary SVM classifiers. Int. J. Approximate Reasoning 72, 55–70 (2016)MathSciNetCrossRefGoogle Scholar
  29. 29.
    Yager, R.R., Liu, L.: Classic Works of the Dempster-Shafer Theory of Belief Functions, vol. 219. Springer, Heidelberg (2008).  https://doi.org/10.1007/978-3-540-44792-4CrossRefzbMATHGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Université de Technologie de Compiègne, CNRS, UMR 7253 HeudiasycCompiègneFrance

Personalised recommendations