Advertisement

Selecting Influential Examples: Active Learning with Expected Model Output Changes

  • Alexander Freytag
  • Erik Rodner
  • Joachim Denzler
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8692)

Abstract

In this paper, we introduce a new general strategy for active learning. The key idea of our approach is to measure the expected change of model outputs, a concept that generalizes previous methods based on expected model change and incorporates the underlying data distribution. For each example of an unlabeled set, the expected change of model predictions is calculated and marginalized over the unknown label. This results in a score for each unlabeled example that can be used for active learning with a broad range of models and learning algorithms. In particular, we show how to derive very efficient active learning methods for Gaussian process regression, which implement this general strategy, and link them to previous methods. We analyze our algorithms and compare them to a broad range of previous active learning strategies in experiments showing that they outperform state-of-the-art on well-established benchmark datasets in the area of visual object recognition.

Keywords

active learning Gaussian processes visual recognition exploration-exploitation trade-off 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Supplementary material

978-3-319-10593-2_37_MOESM1_ESM.pdf (3.6 mb)
Electronic Supplementary Material(3,706 KB)

References

  1. 1.
    Baram, Y., El-Yaniv, R., Luz, K.: Online choice of active learning algorithms. Journal of Machine Learning Research (JMLR) 5, 255–291 (2004)MathSciNetGoogle Scholar
  2. 2.
    Deng, J., Berg, A.C., Li, K., Fei-Fei, L.: What does classifying more than 10,000 image categories tell us? In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol. 6315, pp. 71–84. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  3. 3.
    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: Imagenet: A large-scale hierarchical image database. In: Conference on Computer Vision and Pattern Recognition, CVPR (2009)Google Scholar
  4. 4.
    Ebert, S., Fritz, M., Schiele, B.: Ralf: A reinforced active learning formulation for object class recognition. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3626–3633 (2012)Google Scholar
  5. 5.
    Freytag, A., Rodner, E., Bodesheim, P., Denzler, J.: Rapid uncertainty computation with gaussian processes and histogram intersection kernels. In: Lee, K.M., Matsushita, Y., Rehg, J.M., Hu, Z. (eds.) ACCV 2012, Part II. LNCS, vol. 7725, pp. 511–524. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  6. 6.
    Freytag, A., Rodner, E., Bodesheim, P., Denzler, J.: Labeling examples that matter: Relevance-based active learning with gaussian processes. In: Weickert, J., Hein, M., Schiele, B. (eds.) GCPR 2013. LNCS, vol. 8142, pp. 282–291. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  7. 7.
    Göring, C., Rodner, E., Freytag, A., Denzler, J.: Nonparametric part transfer for fine-grained recognition. In: Conference on Computer Vision and Pattern Recognition, CVPR (accepted for publication, 2014)Google Scholar
  8. 8.
    Griffin, G., Holub, A., Perona, P.: Caltech-256 object category dataset. Tech. Rep. 7694. California Institute of Technology (2007)Google Scholar
  9. 9.
    Hariharan, B., Malik, J., Ramanan, D.: Discriminative decorrelation for clustering and classfication. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part IV. LNCS, vol. 7575, pp. 459–472. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  10. 10.
    Kapoor, A., Grauman, K., Urtasun, R., Darrell, T.: Gaussian processes for object categorization. International Journal of Computer Vision (IJCV) 88, 169–188 (2010)CrossRefGoogle Scholar
  11. 11.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, NIPS (2012)Google Scholar
  12. 12.
    Plackett, R.L.: Some theorems in least squares. Biometrika 37(1/2), 149–157 (1950)CrossRefzbMATHMathSciNetGoogle Scholar
  13. 13.
    Rasmussen, C.E., Williams, C.K.I.: Adaptive Computation and Machine Learning. Adaptive Computation and Machine Learning. The MIT Press, Cambridge (2006)Google Scholar
  14. 14.
    Rodner, E., Freytag, A., Bodesheim, P., Denzler, J.: Large-scale gaussian process classification with flexible adaptive histogram kernels. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part IV. LNCS, vol. 7575, pp. 85–98. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  15. 15.
    Rodner, E., Wacker, E.S., Kemmler, M., Denzler, J.: One-class classification for anomaly detection in wire ropes with gaussian processes in a few lines of code. In: Conference on Machine Vision Applications (MVA), pp. 219–222 (2011)Google Scholar
  16. 16.
    Roy, N., McCallum, A.: Toward optimal active learning through sampling estimation of error reduction. In: International Conference on Machine Learning (ICML), pp. 441–448 (2001)Google Scholar
  17. 17.
    Schohn, G., Cohn, D.: Less is more: Active learning with support vector machines. In: International Conference on Machine Learning (ICML), pp. 839–846 (2000)Google Scholar
  18. 18.
    Schölkopf, B., Smola, A.J.: Learning with kernels: Support Vector Machines, Regularization, Optimization, and beyond. Adaptive Computation and Machine Learning. The MIT Press, Cambridge (2002)Google Scholar
  19. 19.
    Settles, B., Craven, M.: An analysis of active learning strategies for sequence labeling tasks. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1070–1079. Association for Computational Linguistics (2008)Google Scholar
  20. 20.
    Settles, B., Craven, M., Ray, S.: Multiple-instance active learning. In: Advances in Neural Information Processing Systems (NIPS), pp. 1289–1296. MIT Press (2008)Google Scholar
  21. 21.
    Tong, S., Koller, D.: Support vector machine active learning with applications to text classification. Journal of Machine Learning Research (JMLR) 2, 45–66 (2002)zbMATHGoogle Scholar
  22. 22.
    Vezhnevets, A., Buhmann, J.M., Ferrari, V.: Active learning for semantic segmentation with expected change. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3162–3169 (2012)Google Scholar
  23. 23.
    Zhu, X., Lafferty, J., Ghahramani, Z.: Combining active learning and semi-supervised learning using gaussian fields and harmonic functions. In: Workshop on the Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining (ICML-WS), pp. 58–65 (2003)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Alexander Freytag
    • 1
  • Erik Rodner
    • 1
  • Joachim Denzler
    • 1
  1. 1.Computer Vision GroupFriedrich Schiller UniversityJenaGermany

Personalised recommendations