Advertisement

Class-Incremental Domain Adaptation

Conference paper
  • 475 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12358)

Abstract

We introduce a practical Domain Adaptation (DA) paradigm called Class-Incremental Domain Adaptation (CIDA). Existing DA methods tackle domain-shift but are unsuitable for learning novel target-domain classes. Meanwhile, class-incremental (CI) methods enable learning of new classes in absence of source training data, but fail under a domain-shift without labeled supervision. In this work, we effectively identify the limitations of these approaches in the CIDA paradigm. Motivated by theoretical and empirical observations, we propose an effective method, inspired by prototypical networks, that enables classification of target samples into both shared and novel (one-shot) target classes, even under a domain-shift. Our approach yields superior performance as compared to both DA and CI methods in the CIDA paradigm.

Notes

Acknowledgement

This work is supported by a Wipro PhD Fellowship and a grant from Uchhatar Avishkar Yojana (UAY, IISC_010), MHRD, Govt. of India.

Supplementary material

504454_1_En_4_MOESM1_ESM.pdf (1.7 mb)
Supplementary material 1 (pdf 1721 KB)

References

  1. 1.
    Baktashmotlagh, M., Faraki, M., Drummond, T., Salzmann, M.: Learning factorized representations for open-set domain adaptation. In: ICLR (2019)Google Scholar
  2. 2.
    Ben-David, S., Blitzer, J., Crammer, K., Kulesza, A., Pereira, F., Vaughan, J.W.: A theory of learning from different domains. Mach. Learn. 79(1–2), 151–175 (2010)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Ben-David, S., Blitzer, J., Crammer, K., Pereira, F.: Analysis of representations for domain adaptation. In: NeurIPS (2007)Google Scholar
  4. 4.
    Castro, F.M., Marin-Jimenez, M.J., Guil, N., Schmid, C., Alahari, K.: End-to-end incremental learning. In: ECCV (2018)Google Scholar
  5. 5.
    Chang, W.G., You, T., Seo, S., Kwak, S., Han, B.: Domain-specific batch normalization for unsupervised domain adaptation. In: CVPR (2019)Google Scholar
  6. 6.
    Chapelle, O., Zien, A.: Semi-supervised classification by low density separation. In: AISTATS (2005)Google Scholar
  7. 7.
    Dhar, P., Singh, R.V., Peng, K.C., Wu, Z., Chellappa, R.: Learning without memorizing. In: CVPR (2019)Google Scholar
  8. 8.
    Dong, N., Xing, E.P.: Domain adaption in one-shot learning. In: ECML-PKDD (2018)Google Scholar
  9. 9.
    Fort, S.: Gaussian prototypical networks for few-shot learning on omniglot (2017). arXiv preprint arXiv:1708.02735
  10. 10.
    Ganin, Y., Lempitsky, V.: Unsupervised domain adaptation by backpropagation. In: ICML (2015)Google Scholar
  11. 11.
    Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., Lempitsky, V.: Domain-adversarial training of neural networks. JMLR 17(1), 2030–2096 (2016)MathSciNetzbMATHGoogle Scholar
  12. 12.
    Gong, B., Shi, Y., Sha, F., Grauman, K.: Geodesic flow kernel for unsupervised domain adaptation. In: CVPR (2012)Google Scholar
  13. 13.
    Goodfellow, I.J., Mirza, M., Xiao, D., Courville, A., Bengio, Y.: An empirical investigation of catastrophic forgetting in gradient-based neural networks (2013). arXiv preprint arXiv:1312.6211
  14. 14.
    Grandvalet, Y., Bengio, Y.: Semi-supervised learning by entropy minimization. In: NeurIPS (2005)Google Scholar
  15. 15.
    He, K., Zhang, X., Ren, S., Sun, J.: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: ICCV (2015)Google Scholar
  16. 16.
    Hendrycks, D., Gimpel, K.: A baseline for detecting misclassified and out-of-distribution examples in neural networks. In: ICLR (2017)Google Scholar
  17. 17.
    Ioffe, S., Szegedy, C.: Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: ICML (2015)Google Scholar
  18. 18.
    Khosla, A., Zhou, T., Malisiewicz, T., Efros, A.A., Torralba, A.: Undoing the damage of dataset bias. In: ECCV (2012)Google Scholar
  19. 19.
    Kingma, D.P., Ba, J.L.: Adam: A method for stochastic optimization. In: ICLR (2014)Google Scholar
  20. 20.
    Kundu, J.N., Gor, M., Agrawal, D., Babu, R.V.: GAN-Tree: An incrementally learned hierarchical generative framework for multi-modal data distributions. In: ICCV (2019)Google Scholar
  21. 21.
    Kundu, J.N., Lakkakula, N., Babu, R.V.: UM-Adapt: Unsupervised multi-task adaptation using adversarial cross-task distillation. In: ICCV (2019)Google Scholar
  22. 22.
    Kundu, J.N., Uppala, P.K., Pahuja, A., Babu, R.V.: Adadepth: Unsupervised content congruent adaptation for depth estimation. In: CVPR (2018)Google Scholar
  23. 23.
    Kundu, J.N., Venkat, N., Rahul, M.V., Venkatesh Babu, R.: Universal source-free domain adaptation. In: CVPR (2020)Google Scholar
  24. 24.
    Kundu, J.N., Venkat, N., Revanur, A., Rahul, M.V., Venkatesh Babu, R.: Towards inheritable models for open-set domain adaptation. In: CVPR (2020)Google Scholar
  25. 25.
    Kuroki, S., Charoenphakdee, N., Bao, H., Honda, J., Sato, I., Sugiyama, M.: Unsupervised domain adaptation based on source-guided discrepancy. In: AAAI (2019)Google Scholar
  26. 26.
    Lee, D.H.: Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In: Workshop on Challenges in Representation Learning at ICML (2013)Google Scholar
  27. 27.
    Lee, K., Lee, H., Lee, K., Shin, J.: Training confidence-calibrated classifiers for detecting out-of-distribution samples. In: ICLR (2018)Google Scholar
  28. 28.
    Li, Z., Hoiem, D.: Learning without forgetting. TPAMI 40(12), 2935–2947 (2017)CrossRefGoogle Scholar
  29. 29.
    Liu, H., Cao, Z., Long, M., Wang, J., Yang, Q.: Separate to adapt: Open set domain adaptation via progressive separation. In: CVPR (2019)Google Scholar
  30. 30.
    Long, M., Cao, Y., Wang, J., Jordan, M.I.: Learning transferable features with deep adaptation networks. In: ICML (2015)Google Scholar
  31. 31.
    Long, M., Zhu, H., Wang, J., Jordan, M.I.: Unsupervised domain adaptation with residual transfer networks. In: NeurIPS (2016)Google Scholar
  32. 32.
    Lopes, R.G., Fenu, S., Starner, T.: Data-free knowledge distillation for deep neural networks. In: LLD Workshop at NeurIPS (2017)Google Scholar
  33. 33.
    Luo, Z., Zou, Y., Hoffman, J., Fei-Fei, L.F.: Label efficient learning of transferable representations across domains and tasks. In: NeurIPS (2017)Google Scholar
  34. 34.
    Nayak, G.K., Mopuri, K.R., Shaj, V., Radhakrishnan, V.B., Chakraborty, A.: Zero-shot knowledge distillation in deep networks. In: ICML (2019)Google Scholar
  35. 35.
    Pan, S.J., Yang, Q.: A survey on transfer learning. IEEE Trans. Knowl. Data Eng. 22, 1345–1359 (2009)Google Scholar
  36. 36.
    Panareda Busto, P., Gall, J.: Open set domain adaptation. In: ICCV (2017)Google Scholar
  37. 37.
    Pei, Z., Cao, Z., Long, M., Wang, J.: Multi-adversarial domain adaptation. In: AAAI (2018)Google Scholar
  38. 38.
    Peng, H., Li, J., Song, Y., Liu, Y.: Incrementally learning the hierarchical softmax function for neural language models. In: AAAI (2017)Google Scholar
  39. 39.
    Peng, X., Usman, B., Kaushik, N., Hoffman, J., Wang, D., Saenko, K.: Visda: The visual domain adaptation challenge (2017). arXiv preprint arXiv:1710.06924
  40. 40.
    Pereyra, G., Tucker, G., Chorowski, J., Kaiser, Ł., Hinton, G.: Regularizing neural networks by penalizing confident output distributions. In: ICLR (2017)Google Scholar
  41. 41.
    Rebuffi, S.A., Kolesnikov, A., Sperl, G., Lampert, C.H.: iCaRL: Incremental classifier and representation learning. In: CVPR (2017)Google Scholar
  42. 42.
    Ruping, S.: Incremental learning with support vector machines. In: ICDM (2001)Google Scholar
  43. 43.
    Saenko, K., Kulis, B., Fritz, M., Darrell, T.: Adapting visual category models to new domains. In: ECCV (2010)Google Scholar
  44. 44.
    Saito, K., Watanabe, K., Ushiku, Y., Harada, T.: Maximum classifier discrepancy for unsupervised domain adaptation. In: CVPR (2018)Google Scholar
  45. 45.
    Saito, K., Yamamoto, S., Ushiku, Y., Harada, T.: Open set domain adaptation by backpropagation. In: ECCV (2018)Google Scholar
  46. 46.
    Sankaranarayanan, S., Balaji, Y., Castillo, C.D., Chellappa, R.: Generate to adapt: Aligning domains using generative adversarial networks. In: CVPR (2018)Google Scholar
  47. 47.
    Shu, Y., Cao, Z., Long, M., Wang, J.: Transferable curriculum for weakly-supervised domain adaptation. In: AAAI (2019)Google Scholar
  48. 48.
    Snell, J., Swersky, K., Zemel, R.: Prototypical networks for few-shot learning. In: NeurIPS (2017)Google Scholar
  49. 49.
    Sun, B., Saenko, K.: Deep coral: Correlation alignment for deep domain adaptation. In: ECCV Workshops (2016)Google Scholar
  50. 50.
    Torralba, A., Efros, A.A.: Unbiased look at dataset bias. In: CVPR (2011)Google Scholar
  51. 51.
    Tzeng, E., Hoffman, J., Saenko, K., Darrell, T.: Adversarial discriminative domain adaptation. In: CVPR (2017)Google Scholar
  52. 52.
    Tzeng, E., Hoffman, J., Zhang, N., Saenko, K., Darrell, T.: Deep domain confusion: Maximizing for domain invariance (2014). arXiv preprint arXiv:1412.3474
  53. 53.
    Wu, Y., Chen, Y., Wang, L., Ye, Y., Liu, Z., Guo, Y., Fu, Y.: Large scale incremental learning. In: CVPR (2019)Google Scholar
  54. 54.
    You, K., Long, M., Cao, Z., Wang, J., Jordan, M.I.: Universal domain adaptation. In: CVPR (2019)Google Scholar
  55. 55.
    Zheng, Z., Hong, P.: Robust detection of adversarial attacks by modeling the intrinsic properties of deep neural networks. In: NeurIPS (2018)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Video Analytics LabIndian Institute of ScienceBangaloreIndia

Personalised recommendations