Advertisement

ForestHash: Semantic Hashing with Shallow Random Forests and Tiny Convolutional Networks

  • Qiang QiuEmail author
  • José Lezama
  • Alex Bronstein
  • Guillermo Sapiro
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11206)

Abstract

In this paper, we introduce a random forest semantic hashing scheme that embeds tiny convolutional neural networks (CNN) into shallow random forests. A binary hash code for a data point is obtained by a set of decision trees, setting ‘1’ for the visited tree leaf, and ‘0’ for the rest. We propose to first randomly group arriving classes at each tree split node into two groups, obtaining a significantly simplified two-class classification problem that can be a handled with a light-weight CNN weak learner. Code uniqueness is achieved via the random class grouping, whilst code consistency is achieved using a low-rank loss in the CNN weak learners that encourages intra-class compactness for the two random class groups. Finally, we introduce an information-theoretic approach for aggregating codes of individual trees into a single hash code, producing a near-optimal unique hash for each class. The proposed approach significantly outperforms state-of-the-art hashing methods for image retrieval tasks on large-scale public datasets, and is comparable to image classification methods while utilizing a more compact, efficient and scalable representation. This work proposes a principled and robust procedure to train and deploy in parallel an ensemble of light-weight CNNs, instead of simply going deeper.

Notes

Acknowledgements

Work partially supported by AFOSR, ARO, NGA, NSF, ONR. José Lezama was supported by ANII (Uruguay) grant PD_NAC_2015_1_108550.

References

  1. 1.
    Gionis, A., Indyk, P., Motwani, R.: Similarity search in high dimensions via hashing. In: Proceedings of International Conference on Very Large Data Bases (1999)Google Scholar
  2. 2.
    Kulis, B., Grauman, K.: Kernelized locality-sensitive hashing for scalable image search. In: Proceedings of International Conference on Computer vision (2009)Google Scholar
  3. 3.
    Weiss, Y., Torralba, A., Fergus, R.: Spectral hashing. In: Advances in Neural Information Processing Systems (2009)Google Scholar
  4. 4.
    Masci, J., Bronstein, A.M., Bronstein, M.M., Sprechmann, P., Sapiro, G.: Sparse similarity-preserving hashing. In: International Conference on Learning Representations, Banff, Canada, April 2014Google Scholar
  5. 5.
    Liu, W., Wang, J., Ji, R., Jiang, Y., Chang, S.: Supervised hashing with Kernels. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 2012Google Scholar
  6. 6.
    Liu, W., Wang, J., Chang, S.: Hashing with graphs. In: International Conference on Machine Learning (2011)Google Scholar
  7. 7.
    Zhang, D., Wang, J., Cai, D., Lu, J.: Self-taught hashing for fast similarity search. In: Proceedings of International Conference on Research and Development in Information Retrieval (2010)Google Scholar
  8. 8.
    Liu, H., Wang, R., Shan, S., Chen, X.: Deep supervised hashing for fast image retrieval. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2016)Google Scholar
  9. 9.
    Masci, J., Bronstein, M.M., Bronstein, A.M., Schmidhuber, J.: Multimodal similarity-preserving hashing. IEEE Trans. Patt. Anal. Mach. Intell. 36(4), 824–830 (2014)CrossRefGoogle Scholar
  10. 10.
    Norouzi, M., Fleet, D.J., Salakhutdinov, R.: Hamming distance metric learning. In: Advances in Neural Information Processing Systems (2012)Google Scholar
  11. 11.
    Breiman, L.: Random forests. Mach. Learn. 45(1), 5–32 (2001)CrossRefGoogle Scholar
  12. 12.
    Criminisi, A., Shotton, J.: Decision Forests for Computer Vision and Medical Image Analysis. Springer, London (2013).  https://doi.org/10.1007/978-1-4471-4929-3CrossRefGoogle Scholar
  13. 13.
    Shotton, J.: Efficient human pose estimation from single depth images. IEEE Trans. Patt. Anal. Mach. Intell. 35(12), 2821–2840 (2013)CrossRefGoogle Scholar
  14. 14.
    Gall, J., Lempitsky, V.: Class-specific HOUGH forests for object detection. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2009)Google Scholar
  15. 15.
    Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers Inc., Burlington (1993)Google Scholar
  16. 16.
    Fazel, M.: Matrix Rank Minimization with Applications. Ph.D Thesis, Stanford University (2002)Google Scholar
  17. 17.
    Qiu, Q., Sapiro, G.: Learning transformations for clustering and classification. J. Mach. Learn. Res. 16, 187–225 (2015)MathSciNetzbMATHGoogle Scholar
  18. 18.
    Aharon, M., Elad, M., Bruckstein, A.: K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans. Sig. Process. 54(11), 4311–4322 (2006)CrossRefGoogle Scholar
  19. 19.
    Qiu, Q., Sapiro, G.: Learning transformations for classification forests. In: ICLR (2014)Google Scholar
  20. 20.
    Watson, G.A.: Characterization of the subdifferential of some matrix norms. Linear Algebra Appl. 170, 1039–1053 (1992)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Sriperumbudur, B.K., Lanckriet, G.R.G.: A proof of convergence of the concave-convex procedure using Zangwill’s theory. Neural Comput. 24(6), 1391–1407 (2012)MathSciNetCrossRefGoogle Scholar
  22. 22.
    Yuille, A.L., Rangarajan, A.: The concave-convex procedure. Neural Comput. 4, 915–936 (2003)CrossRefGoogle Scholar
  23. 23.
    Lezama, J., Qiu, Q., Musé, P., Sapiro, G.: Olé: orthogonal low-rank embedding, a plug and play geometric loss for deep learning. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2018)Google Scholar
  24. 24.
    Krause, A., Singh, A., Guestrin, C.: Near-optimal sensor placements in Gaussian processes: theory, efficient algorithms and empirical studies. J. Mach. Learn. Res. 9, 235–284 (2008)zbMATHGoogle Scholar
  25. 25.
    Nemhauser, G., Wolsey, L., Fisher, M.: An analysis of approximations for maximizing submodular set functions. Math. Program. 14(1), 265–294 (1978)MathSciNetCrossRefGoogle Scholar
  26. 26.
    Hellman, M.E., Raviv, J.: Probability of error, equivocation, and the Chernoff bound. IEEE Trans. Inf. Theory 16, 368–372 (1979)MathSciNetCrossRefGoogle Scholar
  27. 27.
    Krizhevsky, A.: Learning multiple layers of features from tiny images. Technical report (2009)Google Scholar
  28. 28.
    Lecun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)CrossRefGoogle Scholar
  29. 29.
    Rasiwasia, N., et al.: A new approach to cross-modal multimedia retrieval. In: Proceedings of the International Conference on Multimedia (2010)Google Scholar
  30. 30.
    Gong, Y., Lazebnik, S.: Iterative quantization: a procrustean approach to learning binary codes. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2011)Google Scholar
  31. 31.
    Norouzi, M., Fleet, D.J.: Minimal loss hashing for compact binary codes. In: International Conference on Machine Learning (2011)Google Scholar
  32. 32.
    Kulis, B., Darrell, T.: Learning to hash with binary reconstructive embeddings. In: Advances in Neural Information Processing Systems (2009)Google Scholar
  33. 33.
    Xia, R., Pan, Y., Lai, H., Liu, C., Yan, S.: Supervised hashing for image retrieval via image representation learning. In: Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence (2014)Google Scholar
  34. 34.
    Lin, K., Yang, H.F., Hsiao, J.H., Chen, C.S.: Deep learning of binary hash codes for fast image retrieval. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2015)Google Scholar
  35. 35.
    Lai, H., Pan, Y., Liu, Y., Yan, S.: Simultaneous feature learning and hash coding with deep neural networks. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2015)Google Scholar
  36. 36.
    Lin, M., Chen, Q., Yan, S.: Network In Network. In: ICLR (2014)Google Scholar
  37. 37.
    Springenberg, J.T., Dosovitskiy, A., Brox, T., Riedmiller, M.A.: Striving for simplicity: The all convolutional net. CoRR abs/1412.6806 (2014)Google Scholar
  38. 38.
    Lee, C., Xie, S., Gallagher, P.W., Zhang, Z., Tu, Z.: Deeply-supervised nets. In: AISTATS (2015)Google Scholar
  39. 39.
    Larsson, G., Maire, M., Shakhnarovich, G.: Fractalnet: ultra-deep neural networks without residuals. In: ICLR (2017)Google Scholar
  40. 40.
    Huang, G., Sun, Y., Liu, Z., Sedra, D., Weinberger, K.Q.: Deep networks with stochastic depth. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 646–661. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46493-0_39CrossRefGoogle Scholar
  41. 41.
    He, K., Zhang, X., Ren, S., Sun, J.: Identity mappings in deep residual networks. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 630–645. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46493-0_38CrossRefGoogle Scholar
  42. 42.
    Lin, G., Shen, C., Shi, Q., van den Hengel, A., Suter, D.: Fast supervised hashing with decision trees for high-dimensional data. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2014)Google Scholar
  43. 43.
    Lin, G., Shen, C., Suter, D., van den Hengel, A.: A general two-step approach to learning-based hashing. In: Proceedings of the International Conference on Computer vision (2013)Google Scholar
  44. 44.
    Wang, J., Kumar, S., Chang, S.F.: Sequential projection learning for hashing with compact codes. In: International Conference on Machine Learning, Haifa, Israel (2010)Google Scholar
  45. 45.
    Raginsky, M., Lazebnik, S.: Locality-sensitive binary codes from shift-invariant kernels. In: Advances in Neural Information Processing Systems (2010)Google Scholar
  46. 46.
    Bronstein, M., Bronstein, A., Michel, F., Paragios, N.: Data fusion through cross-modality metric learning using similarity-sensitive hashing. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 2010Google Scholar
  47. 47.
    Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Patt. Anal. Mach. Intell. 35(8), 1798–1828 (2013)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Qiang Qiu
    • 1
    Email author
  • José Lezama
    • 2
  • Alex Bronstein
    • 3
  • Guillermo Sapiro
    • 1
  1. 1.Duke UniversityDurhamUSA
  2. 2.Universidad de la RepúblicaMontevideoUruguay
  3. 3.Technion-Israel Institute of TechnologyHaifaIsrael

Personalised recommendations