Large-Scale Active Learning with Approximations of Expected Model Output Changes

  • Christoph Käding
  • Alexander Freytag
  • Erik Rodner
  • Andrea Perino
  • Joachim Denzler
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9796)

Abstract

Incremental learning of visual concepts is one step towards reaching human capabilities beyond closed-world assumptions. Besides recent progress, it remains one of the fundamental challenges in computer vision and machine learning. Along that path, techniques are needed which allow for actively selecting informative examples from a huge pool of unlabeled images to be annotated by application experts. Whereas a manifold of active learning techniques exists, they commonly suffer from one of two drawbacks: (i) either they do not work reliably on challenging real-world data or (ii) they are kernel-based and not scalable with the magnitudes of data current vision applications need to deal with. Therefore, we present an active learning and discovery approach which can deal with huge collections of unlabeled real-world data. Our approach is based on the expected model output change principle and overcomes previous scalability issues. We present experiments on the large-scale MS-COCO dataset and on a dataset provided by biodiversity researchers. Obtained results reveal that our technique clearly improves accuracy after just a few annotations. At the same time, it outperforms previous active learning approaches in academic and real-world scenarios.

Supplementary material

419026_1_En_15_MOESM1_ESM.pdf (1.6 mb)
Supplementary material 1 (pdf 1622 KB)

References

  1. 1.
    Alajlan, N., Pasolli, E., Melgani, F., Franzoso, A.: Large-scale image classification using active learning. IEEE Geosci. Remote Sens. Lett. 11(1), 259–263 (2014)CrossRefGoogle Scholar
  2. 2.
    Baram, Y., El-Yaniv, R., Luz, K.: Online choice of active learning algorithms. J. Mach. Learn. Res. (JMLR) 5, 255–291 (2004)MathSciNetGoogle Scholar
  3. 3.
    Bishop, C.M.: Pattern Recognition and Machine Learning. Information Science and Statistics. Springer, New York (2006)MATHGoogle Scholar
  4. 4.
    Bouneffouf, D.: Exponentiated gradient exploration for active learning. Computers 5(1), 1 (2016)CrossRefGoogle Scholar
  5. 5.
    Cai, W., Zhang, Y., Zhou, S., Wang, W., Ding, C., Gu, X.: Active learning for support vector machines with maximum model change. In: Calders, T., Esposito, F., Hüllermeier, E., Meo, R. (eds.) ECML PKDD 2014, Part I. LNCS, vol. 8724, pp. 211–226. Springer, Heidelberg (2014)Google Scholar
  6. 6.
    Cohn, D., Atlas, L., Ladner, R.: Improving generalization with active learning. Mach. Learn. 15(2), 201–221 (1994)Google Scholar
  7. 7.
    Demir, B., Bruzzone, L.: A novel active learning method in relevance feedback for content-based remote sensing image retrieval. IEEE Trans. Geosci. Remote Sens. 53(5), 2323–2334 (2015)CrossRefGoogle Scholar
  8. 8.
    Ertekin, S., Huang, J., Bottou, L., Giles, L.: Learning on the border: active learning in imbalanced data classification. In: ACM Conference on Information and Knowledge Management, pp. 127–136 (2007)Google Scholar
  9. 9.
    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
  10. 10.
    Freytag, A., Rodner, E., Denzler, J.: Selecting influential examples: active learning with expected model output changes. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part IV. LNCS, vol. 8692, pp. 562–577. Springer, Heidelberg (2014)Google Scholar
  11. 11.
    Fu, C., Yang, Y.: A batch-mode active learning SVM method based on semi-supervised clustering. Intell. Data Anal. 19(2), 345–358 (2015)Google Scholar
  12. 12.
    Haines, T.S.F., Xiang, T.: Active rare class discovery and classification using dirichlet processes. Int. J. Comput. Vision (IJCV) 106(3), 315–331 (2014)CrossRefGoogle Scholar
  13. 13.
    Hoi, S.C., Jin, R., Lyu, M.R.: Large-scale text categorization by batch mode active learning. In: ACM International Conference on World Wide Web, pp. 633–642 (2006)Google Scholar
  14. 14.
    Huang, S.J., Jin, R., Zhou, Z.H.: Active learning by querying informative and representative examples. In: Neural Information Processing Systems (NIPS), pp. 892–900 (2010)Google Scholar
  15. 15.
    Jain, P., Kapoor, A.: Active learning for large multi-class problems. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 762–769 (2009)Google Scholar
  16. 16.
    Joshi, A., Porikli, F., Papanikolopoulos, N.: Multi-class active learning for image classification. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2372–2379 (2009)Google Scholar
  17. 17.
    Käding, C., Freytag, A., Rodner, E., Bodesheim, P., Denzler, J.: Active learning and discovery of object categories in the presence of unnameable instances. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4343–4352 (2015)Google Scholar
  18. 18.
    Kapoor, A., Grauman, K., Urtasun, R., Darrell, T.: Gaussian processes for object categorization. Int. J. Comput. Vision (IJCV) 88, 169–188 (2010)CrossRefGoogle Scholar
  19. 19.
    Krähenbühl, P., Koltun, V.: Geodesic object proposals. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part V. LNCS, vol. 8693, pp. 725–739. Springer, Heidelberg (2014)Google Scholar
  20. 20.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Neural Information Processing Systems (NIPS), pp. 1097–1105 (2012)Google Scholar
  21. 21.
    Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part V. LNCS, vol. 8693, pp. 740–755. Springer, Heidelberg (2014)Google Scholar
  22. 22.
    Plackett, R.L.: Some theorems in least squares. Biometrika 37(1/2), 149–157 (1950)MathSciNetCrossRefMATHGoogle Scholar
  23. 23.
    Press, W.H.: Numerical Recipes: The Art of Scientific Computing, 3rd edn. Cambridge University Press, Cambridge (2007)MATHGoogle Scholar
  24. 24.
    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
  25. 25.
    Seeger, M.: Low rank updates for the cholesky decomposition. Technical report, University of California, Berkeley (2004)Google Scholar
  26. 26.
    Settles, B.: Active learning literature survey. Computer Sciences Technical Report 1648, University of Wisconsin-Madison (2009)Google Scholar
  27. 27.
    Tong, S., Koller, D.: Support vector machine active learning with applications to text classification. J. Mach. Learn. Res. (JMLR) 2, 45–66 (2002)MATHGoogle Scholar
  28. 28.
    Vezhnevets, A., Buhmann, J.M., Ferrari, V.: Active learning for semantic segmentation with expected change. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2012)Google Scholar
  29. 29.
    Yang, Y., Ma, Z., Nie, F., Chang, X., Hauptmann, A.G.: Multi-class active learning by uncertainty sampling with diversity maximization. Int. J. Comput. Vision (IJCV) 113(2), 113–127 (2014)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  • Christoph Käding
    • 1
    • 2
  • Alexander Freytag
    • 1
    • 2
  • Erik Rodner
    • 1
    • 2
  • Andrea Perino
    • 3
    • 4
  • Joachim Denzler
    • 1
    • 2
    • 3
  1. 1.Computer Vision GroupFriedrich Schiller University JenaJenaGermany
  2. 2.Michael Stifel Center JenaJenaGermany
  3. 3.German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-LeipzigLeipzigGermany
  4. 4.Institute of BiologyMartin Luther University Halle-WittenbergHalle (Saale)Germany

Personalised recommendations