Stochastic Learning of Multi-Instance Dictionary for Earth Mover's Distance based Histogram Comparison

Preprint English OPEN
Fan, Jihong ; Liang, Ru-Ze (2016)
  • Subject: Computer Science - Computer Vision and Pattern Recognition
    arxiv: Computer Science::Machine Learning | Computer Science::Computer Vision and Pattern Recognition
    acm: ComputingMethodologies_PATTERNRECOGNITION | ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION

Dictionary plays an important role in multi-instance data representation. It maps bags of instances to histograms. Earth mover's distance (EMD) is the most effective histogram distance metric for the application of multi-instance retrieval. However, up to now, there is no existing multi-instance dictionary learning methods designed for EMD based histogram comparison. To fill this gap, we develop the first EMD-optimal dictionary learning method using stochastic optimization method. In the stochastic learning framework, we have one triplet of bags, including one basic bag, one positive bag, and one negative bag. These bags are mapped to histograms using a multi-instance dictionary. We argue that the EMD between the basic histogram and the positive histogram should be smaller than that between the basic histogram and the negative histogram. Base on this condition, we design a hinge loss. By minimizing this hinge loss and some regularization terms of the dictionary, we update the dictionary instances. The experiments over multi-instance retrieval applications shows its effectiveness when compared to other dictionary learning methods over the problems of medical image retrieval and natural language relation classification.
  • References (52)
    52 references, page 1 of 6

    1. Beecks, C., Uysal, M., Seidl, T.: Earth mover's distance vs. quadratic form distance: An analytical and empirical comparison. In: Proceedings - 2015 IEEE International Conference on Broadband and Wireless Computing, Communication and Applications, BWCCA 2015, pp. 587-590 (2016)

    5. Chen, Y., Bi, J., Wang, J.Z.: Miles: Multiple-instance learning via embedded instance selection. Pattern Analysis and Machine Intelligence, IEEE Transactions on 28(12), 1931-1947 (2006)

    6. Chen, Y.H., Chen, T.C., Ma, T.C., Lee, T.H., Chen, L.G., et al.: Sub-microwatt knn classifier for implantable closed-loop epileptic neuromodulation system. In: Proceedings of the 2009 International Symposium on Bioelectronics and Bioinformatics, p. 13. RMIT University, School of Electrical and Computer Engineering (2009)

    7. Clarkson, E., Cushing, J.: Shannon information and receiver operating characteristic analysis for multiclass classification in imaging. Journal of the Optical Society of America A: Optics and Image Science, and Vision 33(5), 930-937 (2016)

    8. Fan, X., Malone, B., Yuan, C.: Finding optimal bayesian network structures with constraints learned from data. In: Proceedings of the 30th annual conference on uncertainty in artificial intelligence (UAI-14), pp. 200-209 (2014)

    9. Fan, X., Tang, K.: Enhanced maximum auc linear classifier. In: Fuzzy Systems and Knowledge Discovery (FSKD), 2010 Seventh International Conference on, vol. 4, pp. 1540-1544. IEEE (2010)

    10. Fan, X., Yuan, C.: An improved lower bound for bayesian network structure learning. In: AAAI, pp. 3526-3532 (2015)

    11. Fan, X., Yuan, C., Malone, B.M.: Tightening bounds for bayesian network structure learning. In: AAAI, pp. 2439-2445. Citeseer (2014)

    12. Fu, Z., Robles-Kelly, A., Zhou, J.: Milis: Multiple instance learning with instance selection. Pattern Analysis and Machine Intelligence, IEEE Transactions on 33(5), 958-977 (2011)

    13. Goadrich, M., Oliphant, L., Shavlik, J.: Gleaner: Creating ensembles of first-order clauses to improve recall-precision curves. Machine Learning 64(1-3), 231-261 (2006)

  • Metrics
    No metrics available
Share - Bookmark