Advertisement

Gabor Layers Enhance Network Robustness

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12354)

Abstract

We revisit the benefits of merging classical vision concepts with deep learning models. In particular, we explore the effect of replacing the first layers of various deep architectures with Gabor layers (i.e. convolutional layers with filters that are based on learnable Gabor parameters) on robustness against adversarial attacks. We observe that architectures with Gabor layers gain a consistent boost in robustness over regular models and maintain high generalizing test performance. We then exploit the analytical expression of Gabor filters to derive a compact expression for a Lipschitz constant of such filters, and harness this theoretical result to develop a regularizer we use during training to further enhance network robustness. We conduct extensive experiments with various architectures (LeNet, AlexNet, VGG16, and WideResNet) on several datasets (MNIST, SVHN, CIFAR10 and CIFAR100) and demonstrate large empirical robustness gains. Furthermore, we experimentally show how our regularizer provides consistent robustness improvements.

Keywords

Gabor Robustness Adversarial attacks Regularizer 

Notes

Acknowledgements

This work was partially supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. OSR-CRG2019-4033.

Supplementary material

504446_1_En_26_MOESM1_ESM.pdf (750 kb)
Supplementary material 1 (pdf 749 KB)

References

  1. 1.
    Alekseev, A., Bobe, A.: Gabornet: gabor filters with learnable parameters in deep convolutional neural network. In: International Conference on Engineering and Telecommunication (EnT) (2019)Google Scholar
  2. 2.
    Atanov, A., Ashukha, A., Struminsky, K., Vetrov, D., Welling, M.: The deep weight prior. In: International Conference on Learning Representations (ICLR) (2019)Google Scholar
  3. 3.
    Bai, M., Urtasun, R.: Deep watershed transform for instance segmentation. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)Google Scholar
  4. 4.
    Bibi, A., Ghanem, B., Koltun, V., Ranftl, R.: Deep layers as stochastic solvers. In: International Conference on Learning Representations (ICLR) (2019)Google Scholar
  5. 5.
    Cao, Y., et al.: Adversarial sensor attack on lidar-based perception in autonomous driving. In: ACM SIGSAC Conference on Computer and Communications Security (2019)Google Scholar
  6. 6.
    Carlini, N., Wagner, D.: Towards evaluating the robustness of neural networks. In: IEEE Symposium on Security and Privacy (SP) (2017)Google Scholar
  7. 7.
    Chen, Y., Zhu, L., Ghamisi, P., Jia, X., Li, G., Tang, L.: Hyperspectral images classification with gabor filtering and convolutional neural network. IEEE Geosci. Remote Sens. Lett. 14, 2355–2359 (2017)CrossRefGoogle Scholar
  8. 8.
    Liu, C., Wechsler, H.: Independent component analysis of gabor features for face recognition. IEEE Trans. Neural Netw. 14, 919–928 (2003) CrossRefGoogle Scholar
  9. 9.
    Chernikova, A., Oprea, A., Nita-Rotaru, C., Kim, B.: Are self-driving cars secure? evasion attacks against deep neural networks for steering angle prediction. In: IEEE Security and Privacy Workshops (SPW) (2019)Google Scholar
  10. 10.
    Chollet, F.: Xception: deep learning with depthwise separable convolutions. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)Google Scholar
  11. 11.
    Cisse, M., Bojanowski, P., Grave, E., Dauphin, Y., Usunier, N.: Parseval networks: improving robustness to adversarial examples. In: International Conference on Machine Learning (ICML) (2017)Google Scholar
  12. 12.
    Farabet, C., Couprie, C., Najman, L., LeCun, Y.: Learning hierarchical features for scene labeling. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1915–1929 (2012)CrossRefGoogle Scholar
  13. 13.
    Gabor, D.: Theory of communication. Part 1: The analysis of information. J. Inst. Electr. Eng. Part III Radio Commun. Eng. 93, 429–441 (1946)Google Scholar
  14. 14.
    Goldstein, T., Osher, S.: The split Bregman method for L1-regularized problems. SIAM J. Imag. Sci. 2, 323–343 (2009)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining and harnessing adversarial examples. In: International Conference on Learning Representations (ICLR) (2015)Google Scholar
  16. 16.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Patter Recognition (CVPR) (2016)Google Scholar
  17. 17.
    Hinton, G., et al.: Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Sig. Process. Mag. 29, 82–97 (2012)CrossRefGoogle Scholar
  18. 18.
    Hubel, D.H., Wiesel, T.N.: Receptive fields of single neurons in the cat’s striate cortex. J. Physiol. 148, 574–591 (1959)CrossRefGoogle Scholar
  19. 19.
    Julesz, B.: Textons, the elements of texture perception, and their interactions. Nature 290, 91–97 (1981)CrossRefGoogle Scholar
  20. 20.
    Krizhevsky, A., Hinton, G.: Learning multiple layers of features from tiny images. Technical report, Citeseer (2009)Google Scholar
  21. 21.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Neural Information Processing Systems (NeurIPS) (2012)Google Scholar
  22. 22.
    LeCun, Y.: The MNIST database of handwritten digits (1998). http://yann.lecun.com/exdb/mnist/
  23. 23.
    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521, 436–444 (2015)CrossRefGoogle Scholar
  24. 24.
    LeCun, Y., Haffner, P., Bottou, L., Bengio, Y.: Object recognition with gradient-based learning. Shape, Contour and Grouping in Computer Vision. LNCS, vol. 1681, pp. 319–345. Springer, Heidelberg (1999).  https://doi.org/10.1007/3-540-46805-6_19CrossRefGoogle Scholar
  25. 25.
    Leung, T., Malik, J.: Representing and recognizing the visual appearance of materials using three-dimensional textons. Int. J. Comput. Vis. (IJCV) 43, 29–44 (2001)CrossRefGoogle Scholar
  26. 26.
    Lowe, D.G.: Object recognition from local scale-invariant features. In: International Conference on Computer Vision (ICCV). IEEE (1999)Google Scholar
  27. 27.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. (IJCV) 60, 91–110 (2004)CrossRefGoogle Scholar
  28. 28.
    Luan, S., Chen, C., Zhang, B., Han, J., Liu, J.: Gabor convolutional networks. IEEE Trans. Image Process. 27, 4357–4366 (2018)MathSciNetCrossRefGoogle Scholar
  29. 29.
    Madry, A., Makelov, A., Schmidt, L., Tsipras, D., Vladu, A.: Towards deep learning models resistant to adversarial attacks. In: International Conference on Learning Representations (ICLR) (2018)Google Scholar
  30. 30.
    Marr, D.: Vision: a computational investigation into the human representation and processing of visual information. In: PsycCRITIQUES (1982)Google Scholar
  31. 31.
    Montufar, G., Pascanu, R., Cho, K., Bengio, Y.: On the number of linear regions of deep neural networks. In: Advances in Neural Information Processing Systems (NeurIPS) (2014)Google Scholar
  32. 32.
    Moosavi-Dezfooli, S.M., Fawzi, A., Frossard, P.: Deepfool: a simple and accurate method to fool deep neural networks. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)Google Scholar
  33. 33.
    Namuduri, K.R., Mehrotra, R., Ranganathan, N.: Edge detection models based on Gabor filters. In: International Conference on Pattern Recognition. Conference C: Image, Speech and Signal Analysis (1992)Google Scholar
  34. 34.
    Novak, C.L., Shafer, S.A., et al.: Anatomy of a color histogram. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (1992)Google Scholar
  35. 35.
    Ouyang, W., Wang, X.: Joint deep learning for pedestrian detection. In: IEEE International Conference on Computer Vision (ICCV) (2013)Google Scholar
  36. 36.
    Papernot, N., McDaniel, P., Wu, X., Jha, S., Swami, A.: Distillation as a defense to adversarial perturbations against deep neural networks. In: IEEE Symposium on Security and Privacy (SP) (2016)Google Scholar
  37. 37.
    Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell. 12, 629–639 (1990)CrossRefGoogle Scholar
  38. 38.
    Poggi, M., Mattoccia, S.: A wearable mobility aid for the visually impaired based on embedded 3D vision and deep learning. In: 2016 IEEE Symposium on Computers and Communication (ISCC) (2016)Google Scholar
  39. 39.
    Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagating errors. Nature 323, 533–536 (1986)CrossRefGoogle Scholar
  40. 40.
    Russakovsky, O., et al.: Imagenet large scale visual recognition challenge. In: International Journal of Computer Vision (IJCV) (2015)Google Scholar
  41. 41.
    Sarwar, S.S., Panda, P., Roy, K.: Gabor filter assisted energy efficient fast learning convolutional neural networks. CoRR (2017)Google Scholar
  42. 42.
    Sedghi, H., Gupta, V., Long, P.M.: The singular values of convolutional layers. In: International Conference on Learning Representations (ICLR) (2019)Google Scholar
  43. 43.
    Shafahi, A., et al.: Adversarial training for free! In: Neural Information Processing Systems (NeurIPS) (2019)Google Scholar
  44. 44.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: International Conference on Learning Representations (ICLR) (2015)Google Scholar
  45. 45.
    Su, Y.M., Wang, J.F.: A novel stroke extraction method for Chinese characters using Gabor filters. Pattern Recogn. 36, 635–647 (2003)CrossRefGoogle Scholar
  46. 46.
    Sun, J., Li, H., Xu, Z., et al.: Deep ADMM-net for compressive sensing MRI. In: Neural Information Systems (NeurIPS) (2016)Google Scholar
  47. 47.
    Szegedy, C., et al.: Intriguing properties of neural networks. In: International Conference on Learning Representations (ICLR) (2014)Google Scholar
  48. 48.
    Wong, E., Rice, L., Kolter, J.Z.: Fast is better than free: revisiting adversarial training. In: International Conference on Learning Representations (ICLR) (2020)Google Scholar
  49. 49.
    Xu, K., et al.: Structured adversarial attack: towards general implementation and better interpretability. In: International Conference on Learning Representations (ICLR) (2019)Google Scholar
  50. 50.
    Yao, H., Chuyi, L., Dan, H., Weiyu, Y.: Gabor feature based convolutional neural network for object recognition in natural scene. In: International Conference on Information Science and Control Engineering (ICISCE) (2016)Google Scholar
  51. 51.
    Zagoruyko, S., Komodakis, N.: Wide residual networks. In: British Machine Vision Conference (BMVC) (2016)Google Scholar
  52. 52.
    Zhang, D., Zhang, T., Lu, Y., Zhu, Z., Dong, B.: You only propagate once: accelerating adversarial training via maximal principle. In: Neural Information Processing Systems (NeurIPS) (2019)Google Scholar
  53. 53.
    Zhang, J., Ghanem, B.: Deep learning. In: IEEE Conference on Computer Vision and Patter Recognition (CVPR) (2017)Google Scholar
  54. 54.
    Zhang, R.: Making convolutional networks shift-invariant again. In: International Con-ference on Machine Learning (ICML) (2019)Google Scholar
  55. 55.
    Zhong, Z., Jin, L., Xie, Z.: High performance offline handwritten Chinese character recognition using Googlenet and directional feature maps. In: International Conference on Document Analysis and Recognition (ICDAR) (2015)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Center for Research and Formation in Artificial IntelligenceUniversidad de los AndesBogotaColombia
  2. 2.King Abdullah University of Science and Technology (KAUST)ThuwalSaudi Arabia

Personalised recommendations