Skip to main content

2022 | OriginalPaper | Buchkapitel

Grad\(_{2}\)VAE: An Explainable Variational Autoencoder Model Based on Online Attentions Preserving Curvatures of Representations

verfasst von : Mohanad Abukmeil, Stefano Ferrari, Angelo Genovese, Vincenzo Piuri, Fabio Scotti

Erschienen in: Image Analysis and Processing – ICIAP 2022

Verlag: Springer International Publishing

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

Unsupervised learning (UL) is a class of machine learning (ML) that learns data, reduces dimensionality, and visualizes decisions without labels. Among UL models, a variational autoencoder (VAE) is considered a UL model that is regulated by variational inference to approximate the posterior distribution of large datasets. In this paper, we propose a novel explainable artificial intelligence (XAI) method to visually explain the VAE behavior based on the second-order derivative of the latent space concerning the encoding layers, which reflects the amount of acceleration required from encoding to decoding space. Our model is termed as Grad\(_{2}\)VAE and it is able to capture the local curvatures of the representations to build online attention that visually explains the model’s behavior. Besides the VAE explanation, we employ our method for anomaly detection, where our model outperforms the recent UL deep models when generalizing it for large-scale anomaly data.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Literatur
1.
Zurück zum Zitat Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: On approximating the non-negative rank: applications to image reduction. In: Proceedings of CIVEMSA (2020) Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: On approximating the non-negative rank: applications to image reduction. In: Proceedings of CIVEMSA (2020)
2.
Zurück zum Zitat Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: Unsupervised learning from limited available data by \(\beta \)-NMF and dual autoencoder. In: Proceedings of ICIP (2020) Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: Unsupervised learning from limited available data by \(\beta \)-NMF and dual autoencoder. In: Proceedings of ICIP (2020)
3.
Zurück zum Zitat Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: A survey of unsupervised generative models for exploratory data analysis and representation learning. ACM Comput. Surv. (CSUR) 54(5), 1–40 (2021)CrossRef Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., Scotti, F.: A survey of unsupervised generative models for exploratory data analysis and representation learning. ACM Comput. Surv. (CSUR) 54(5), 1–40 (2021)CrossRef
4.
Zurück zum Zitat Abukmeil, M., Genovese, A., Piuri, V., Rundo, F., Scotti, F.: Towards explainable semantic segmentation for autonomous driving systems by multi-scale variational attention. In: Proceedings of ICAS (2021) Abukmeil, M., Genovese, A., Piuri, V., Rundo, F., Scotti, F.: Towards explainable semantic segmentation for autonomous driving systems by multi-scale variational attention. In: Proceedings of ICAS (2021)
5.
Zurück zum Zitat Ames, W.F.: Numerical Methods for Partial Differential Equations. Academic Press, Cambridge (2014) Ames, W.F.: Numerical Methods for Partial Differential Equations. Academic Press, Cambridge (2014)
6.
Zurück zum Zitat Baldi, P.: Autoencoders, unsupervised learning, and deep architectures. In: Proceedings of ICML (2012) Baldi, P.: Autoencoders, unsupervised learning, and deep architectures. In: Proceedings of ICML (2012)
7.
Zurück zum Zitat Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013) Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)
8.
Zurück zum Zitat Fan, K., Wang, Z., Beck, J., Kwok, J., Heller, K.A.: Fast second order stochastic backpropagation for variational inference. In: Proceedings of NIPS (2015) Fan, K., Wang, Z., Beck, J., Kwok, J., Heller, K.A.: Fast second order stochastic backpropagation for variational inference. In: Proceedings of NIPS (2015)
9.
Zurück zum Zitat Genovese, A., Piuri, V., Scotti, F.: Towards explainable face aging with generative adversarial networks. In: Proceedings of ICIP (2019) Genovese, A., Piuri, V., Scotti, F.: Towards explainable face aging with generative adversarial networks. In: Proceedings of ICIP (2019)
10.
Zurück zum Zitat He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of CVPR (2016) He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of CVPR (2016)
11.
Zurück zum Zitat Hinton, G.E., Salakhutdinov, R.R.: Reducing the dimensionality of data with neural networks. Science 313(5786), 504–507 (2006)MathSciNetCrossRef Hinton, G.E., Salakhutdinov, R.R.: Reducing the dimensionality of data with neural networks. Science 313(5786), 504–507 (2006)MathSciNetCrossRef
12.
Zurück zum Zitat Hinton, G.E., Zemel, R.S.: Autoencoders, minimum description length and Helmholtz free energy. In: Proceedings of NIPS (1994) Hinton, G.E., Zemel, R.S.: Autoencoders, minimum description length and Helmholtz free energy. In: Proceedings of NIPS (1994)
13.
Zurück zum Zitat Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of ICML (2014) Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of ICML (2014)
14.
Zurück zum Zitat Kingma, D.P., Welling, M.: Auto-encoding variational Bayes. In: Proceedings of ICLR (2014) Kingma, D.P., Welling, M.: Auto-encoding variational Bayes. In: Proceedings of ICLR (2014)
16.
Zurück zum Zitat Liu, W., et al.: Towards visually explaining variational autoencoders. In: Proceedings of CVPR (2020) Liu, W., et al.: Towards visually explaining variational autoencoders. In: Proceedings of CVPR (2020)
17.
18.
Zurück zum Zitat Mejjati, Y.A., Richardt, C., Tompkin, J., Cosker, D., Kim, K.I.: Unsupervised attention-guided image-to-image translation. In: Proceedings of NIPS (2018) Mejjati, Y.A., Richardt, C., Tompkin, J., Cosker, D., Kim, K.I.: Unsupervised attention-guided image-to-image translation. In: Proceedings of NIPS (2018)
19.
Zurück zum Zitat Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. In: Proceedings of ICML (2010) Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. In: Proceedings of ICML (2010)
20.
Zurück zum Zitat Ng, A.: Sparse autoencoder. CS294A Lecture Notes, vol. 72, no. 2011, pp. 1–19 (2011) Ng, A.: Sparse autoencoder. CS294A Lecture Notes, vol. 72, no. 2011, pp. 1–19 (2011)
22.
Zurück zum Zitat Rezende, D.J., Mohamed, S., Wierstra, D.: Stochastic backpropagation and approximate inference in deep generative models. In: Proceedings of ICML (2014) Rezende, D.J., Mohamed, S., Wierstra, D.: Stochastic backpropagation and approximate inference in deep generative models. In: Proceedings of ICML (2014)
23.
Zurück zum Zitat Rifai, S., Vincent, P., Muller, X., Glorot, X., Bengio, Y.: Contractive auto-encoders: explicit invariance during feature extraction. In: Proceedings of ICML (2011) Rifai, S., Vincent, P., Muller, X., Glorot, X., Bengio, Y.: Contractive auto-encoders: explicit invariance during feature extraction. In: Proceedings of ICML (2011)
24.
Zurück zum Zitat Ruff, L., et al.: Deep one-class classification. In: Proceedings of ICML (2018) Ruff, L., et al.: Deep one-class classification. In: Proceedings of ICML (2018)
25.
Zurück zum Zitat Sarafijanovic-Djukic, N., Davis, J.: Fast distance-based anomaly detection in images using an inception-like autoencoder. In: Proceedings of DS (2019) Sarafijanovic-Djukic, N., Davis, J.: Fast distance-based anomaly detection in images using an inception-like autoencoder. In: Proceedings of DS (2019)
26.
Zurück zum Zitat Tang, C., Srivastava, N., Salakhutdinov, R.R.: Learning generative models with visual attention. In: Proceedings of NIPS (2014) Tang, C., Srivastava, N., Salakhutdinov, R.R.: Learning generative models with visual attention. In: Proceedings of NIPS (2014)
27.
Zurück zum Zitat Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: Proceedings of ICML (2008) Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: Proceedings of ICML (2008)
28.
Zurück zum Zitat Xiao, H., Rasul, K., Vollgraf, R.: Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747 (2017) Xiao, H., Rasul, K., Vollgraf, R.: Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:​1708.​07747 (2017)
31.
Zurück zum Zitat Zhang, C., Butepage, J., Kjellstrom, H., Mandt, S.: Advances in variational inference. IEEE Trans. Pattern Anal. Mach. Intell. 41(08), 2008–2026 (2019) Zhang, C., Butepage, J., Kjellstrom, H., Mandt, S.: Advances in variational inference. IEEE Trans. Pattern Anal. Mach. Intell. 41(08), 2008–2026 (2019)
32.
Zurück zum Zitat Zhang, J., Bargal, S.A., Lin, Z., Brandt, J., Shen, X., Sclaroff, S.: Top-down neural attention by excitation backprop. Int. J. Comput. Vis. 126(10), 1084–1102 (2018)CrossRef Zhang, J., Bargal, S.A., Lin, Z., Brandt, J., Shen, X., Sclaroff, S.: Top-down neural attention by excitation backprop. Int. J. Comput. Vis. 126(10), 1084–1102 (2018)CrossRef
Metadaten
Titel
GradVAE: An Explainable Variational Autoencoder Model Based on Online Attentions Preserving Curvatures of Representations
verfasst von
Mohanad Abukmeil
Stefano Ferrari
Angelo Genovese
Vincenzo Piuri
Fabio Scotti
Copyright-Jahr
2022
DOI
https://doi.org/10.1007/978-3-031-06427-2_56

Premium Partner