Skip to main content
SpringerLink
Log in

Few-shot learning based on enhanced pseudo-labels and graded pseudo-labeled data selection

  • Original Article
  • Published:
International Journal of Machine Learning and Cybernetics Aims and scope Submit manuscript

Abstract

Pseudo-labeled data is used to solve the data shortage in few-shot learning, in which the quality of pseudo-labels and pseudo-labeled data selection determine the classification performance. In order to obtain the enhanced pseudo-labels, we used diverse inputs to encourage the label network to learn invariant and robust representations, improving the generalization ability. Simultaneously, the depthwise over-parameterized convolutional layer and group residual connection with shared parameters accelerate the network training and overcome the time-consuming caused by diverse inputs. Then, the graded pseudo-labeled data selection is proposed to determine various quantities of pseudo-labeled data based on the label network’s performance level, which improves the classification accuracy and avoids the high consumption caused by using all the pseudo-labeled data. Finally, we solved the data shortage in food recognition with the proposed method. The experiments show that our method has better classification accuracy and generalization ability in few-shot benchmark datasets and food recognition with few samples.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Data availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Gao F, Cai L, Yang Z, Song S, Wu C (2022) Multi-distance metric network for few-shot learning. Int J Mach Learn Cybern 13(9):2495–2506

    Article  Google Scholar 

  2. Wang K, Wang X, Zhang T, Cheng Y (2022) Few-shot learning with deep balanced network and acceleration strategy. Int J Mach Learn Cybern 13(1):133–144

    Article  Google Scholar 

  3. Finn C, Abbeel P, Levine S (2017) Model-agnostic meta-learning for fast adaptation of deep networks. In: Proceedings of the International Conference on Machine Learning (ICML), Sydney, AUSTRALIA, pp 1126–1135

  4. Nichol A, Achiam J, Schulman J (2018) On first-order meta-learning algorithms. arXiv:1803.02999

  5. Rusu AA, Rao D, Sygnowski J, et al (2018) Meta-learning with latent embedding optimization. arXiv:1807.05960

  6. Santoro A, Bartunov S, Botvinick M, Wierstra D, Lillicrap T (2016) Meta-learning with memory-augmented neural netowrks. In: Proceedings of the International Conference on Machine Learning (ICML), New York City, NY, USA, pp 1842–1850

  7. Munkhdalai T, Yuan X, Mehri, S et al (2018) Rapid adaptation with conditionally shifted neurons. In: Proceedings of the International Conference on Machine Learning (ICML), Stockholm, SWEDEN, pp 3661–3670

  8. Ravi S, Larochelle H (2017) Optimization as a model for few-shot learning. In: Proceedings of the International Conference on Learning Representations (ICLR), Toulon, FRANCE, https://openreview.net/forum?id=rJY0-Kcll

  9. Li S, Chen D, Liu B, et al (2019) Memory-based neighbourhood embedding for visual recognition. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), Seoul, Koera (South), pp 6101–6110

  10. Vinyals O, Blundell C, Lillicrap T, et al (2016) Matching networks for one shot learning. In: Proceedings of the Advances in Neural Information Processing Systems (NIPS), Barcelona, SPAIN, pp 3630–3638

  11. Snell J, Swersky K, Zemel RS (2017) Prototypical networks for few-shot learning. In: Proceedings of the Advances in Neural Information Processing Systems (NIPS), Long Beach, CA, USA, pp 4077–4087

  12. Sung F, Yang Y, Zhang L, et al (2018) Learning to compare: relation network for few-shot learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, pp 1199–1208

  13. Oreshkin BN, Rodriguez P, Lacoste A (2018) Tadam: task dependent adaptive metric for improved few-shot learning. In: Proceedings of the Advances in Neural Information Processing Systems (NIPS), Montréal , CANADA, pp 719–729

  14. Antoniou A, Storkey A, Edwards H (2017) Data augmentation generative adversarial networks. arXiv: 1711.04340

  15. Zhang R, Che T, Ghahramani Z et al (2018) Metagan: an adversarial approach to few-shot learning. In: Proceedings of the Advances in Neural Information Processing Systems (NIPS), Montréal, CANADA, pp 2371–2380

  16. Xu W, Guo D, Qian Y, Ding W (2022) Two-way concept-cognitive learning method: a fuzzy-based progressive learning. IEEE Trans Fuzzy Syst. https://doi.org/10.1109/TFUZZ.2022.3216110

    Article  Google Scholar 

  17. Xu W, Li W (2016) Granular computing approach to two-way learning based on formal concept analysis in fuzzy datasets. IEEE Trans Cybern 46(2):366–379

    Article  MathSciNet  Google Scholar 

  18. Li K, Zhang Y, Li K, Fu, Y (2020) Adversarial feature hallucination networks for few-shot learning. arXiv: 2003.13193

  19. Kim J, Kim H, Kim G (2020) Model-agnostic boundary-adversarial sampling for test-time generalization in few-shot learning. In: Proceedings of the European Conference on Computer Vision (ECCV), ELECTR NETWORK, pp 599–6175

  20. Chen Z, Fu Y, Wang Y, Ma L, Liu W, Hebert M (2019) Image deformation meta-networks for one-shot learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, pp 8672–8681

  21. Zhang H, Zhang J, Koniusz P (2019) Few-shot learning via saliency-guided hallucination of samples. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, pp 2765–2774

  22. Wang Y, Xu C, Liu C, Zhang L, Fu Y (2020) Instance credibility inference for few-shot learning. arXiv: 2003.11853

  23. Lee K, Lee K, Shin J, Lee H (2019) Network randomization: a simple technique for generalization in deep reinforcement learning. arXiv: 1910.05396v3

  24. Cao J, Li Y, Sun M, et al (2020) DO-Conv: depthwise over-parameterized convolutional layer. arXiv: 2006.12030

  25. Wang X, Yu S (2020) Tied block convolution: leaner and better CNNs with shared thinner filters. arXiv: 2009.12021

  26. Bertinetto L, Henriques J, Torr P, Vedaldi A (2018) Meta-learning with differentiable closed-form solvers. arXiv: 1805.08136

  27. Ren M, Triantafillou E, Ravi S, Snell J, Swersky K, Tenenbaum J, Larochelle H, Zemel R (2018) Meta-learning for semi-supervised few-shot classification. arXiv: 1803.00676

  28. Liu Y, Lee J, Park M, Kim S, Yang E, Hwang S, Yang L (2018) Learning to propagate labels: transductive propagation network for few-shot learning. arXiv: 1805.10002

  29. Simon C, Koniusz P, Nock R, Harandi M (2020) Adaptive subspaces for few-shot learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ELECTR NETWORK, pp 4135–4144

  30. Lee H, Hwang S, Shin J (2019) Self-supervised label augmentation via input transformations. arXiv: 1910.05872

  31. Khrulkov V, Mirvakhabova L, Ustinova E, Oseledets I, Lempitsky V. (2019) Hyperbolic image embeddings. arXiv: 1904.02239

  32. Luo Q, Wang L, Lv J, Xiang S, Pan C (2021) Few-shot learning via feature hallucination with variational inference. In: Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), ELECTR NETWORK, pp 3962–3971

  33. Majumder O, Ravichandran A, Maji S, Polito M, Soatto S (2021) Revisiting contrastive learning for few-shot classification. arXiv: 2101.11058

  34. Chen W, Liu Y, Kira Z, Wang Y, Huang J (2019) A closer look at few-shot classification. arXiv: 1904.04232

  35. Zhang Y, Huang S, Peng X, Yang D (2021) Dizygotic conditional variational autoencoder for multi-modal and partial modality absent few-shot learning. arXiv: 2106.14467

  36. Kang D, Kwon H, Min J, Cho M (2021) Relational embedding for few-shot classification. arXiv: 2108.09666

  37. Afrasiyabi A, Lalonde JF, Gagné C (2020) Mixture-based feature space learning for few-shot image classification. arXiv: 2011.11872

  38. Ye H, Ming L, Zhan D, Chao W (2021) Few-shot learning with a strong teacher. arXiv: 2107.00197

  39. Yuan M, Wang W, Wang T, Cai C, Xu Q, Lu, T (2021) Learning class-level prototypes for few-shot learning. arXiv: 2108.11072

  40. Dhillon G, Chaudhari P, Ravichandran A, Soatto S (2019) A baseline for few-shot image classification. arXiv: 1909.02729

  41. Lee E, Huang C, Lee C. (2021) Few-shot and continual learning with attentive independent mechanisms. arXiv: 2107.14053

  42. Schwarcz S, Rambhatla S, Chellappa R (2021) Self-denoising neural networks for few shot learning. arXiv: 2110.13386

  43. Li W, Wang L, Huo J, Shi Y, Gao Y, Luo J (2020) Asymmetric distribution measure for few-shot learning. arXiv: 2002.00153

  44. Yu Z, Raschka S (2020) Looking back to lower-level information in few-shot learning. arXiv: 2005.13638

  45. Lee K, Maij S, Ravichandran A, Soatto S (2019) Meta-learning with differentiable convex optimization. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, pp 10649–10657

  46. Ren M, Liao R, Fetaya E, Zemel R. (2018) Incremental few-shot learning with attention attractor networks. arXiv: 1810.07218

  47. Qiao L, Shi Y, Li J, Tian Y, Huang T, Wang Y (2019) Transductive episodic-wise adaptive metric for few-shot learning. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), Seoul, Korea (South), pp 3602–3611

  48. Ravichandran A, Bhotika R, Soatto S (2019) Few-shot learning with embedded class models and shot-free meta training. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), Seoul, Korea (South), pp 331–339

  49. Tian Y, Wang Y, Krishnan D, Tenenbau J, Isola P (2020) Rethinking few-shot image classification: a good embedding is all you need? In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ELECTR NETWORK, pp 266–282

  50. Xu W, Xu Y, Wang H, Tu Z (2021) Attentional constellation nets for few-shot learning. In: Proceedings of the International Conference on Learning Representations (ICLR), ELECTR NETWORK, https://openreview.net/forum?id=vujTf_I8Kmc

  51. Lazarou M, Avrithis Y, Stathaki T (2021) Few-shot learning via tensor hallucination. arXiv: 2104.09467

  52. Sun Q, Liu Y, Chen Z, Tat-Seng C, Bernt S (2022) Meta-transfer learning through hard tasks. IEEE Trans Pattern Anal Mach Intell 44(3):1443–1456

    Article  Google Scholar 

  53. Lifchitz Y, Avrithis Y, Picard S, Bursuc A (2019) Dense classification and implanting for few-shot learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, pp 9250–9259

  54. Afrasiyabi A, Lalonde JF, Gagne C (2020) Associative alignment for few-shot image classification. In: Proceedings of the European Conference on Computer Vision (ECCV), ELECTR NETWORK, pp 18–35

  55. Wah C, Branson S, Welinder P, Perona P, Belongie S (2011) “The caltech-ucsd birds-200–2011 dataset. Computation& Neural Systems Technical Report. CNS-TR-2011–001

  56. Hilliard N, Phillips L, Howland S, Yankov A, Corley C, Hodas N (2018) Few-shot learning with metric-agnostic conditional embeddings. arXiv: 1802.04376

  57. Patacchiola M, Turner J, Crowley E, O’Boyle M, Storkey A (2019) Bayesian meta-learning for the few-shot setting via deep kernels. arXiv: 1910.05199

  58. Lazarou M, Avrithis Y, Stathaki T (2021) Tensor feature hallucination for few-shot learning. arXiv: 2106.05321

  59. Bossard L, Guillaumin M, Gool L (2014) Food-101–mining discriminative components withrandom forests,” In: Proceedings of the European Conference on Computer Vision (ECCV), Zurich, SWITZERLAND, pp 446–461

  60. Chen X, Zhu Y, Zhou H, Diao L, Wang D (2017) ChineseFoodNet: a large-scale image dataset for chinese food recognition. arXiv: 1705.02743

  61. Jiang S, Min W, Lv Y, Liu L (2020) Few-shot food recognition via multi-view representation learning. ACM Trans Multimed Comput Commun Appl 16(3):1–20

    Article  Google Scholar 

Download references

Funding

This work was supported by the National Natural Science Foundation of China under Grant 62176259 and Grant 61976215.

Author information

Authors and Affiliations

  1. Engineering Research Center of Intelligent Control for Underground Space, Ministry of Education, China University of Mining and Technology, Xuzhou, 221116, China

    Kang Wang, Xuesong Wang & Yuhu Cheng

  2. School of Information and Control Engineering, China University of Mining and Technology, Xuzhou, 221116, China

    Kang Wang, Xuesong Wang & Yuhu Cheng

Authors
  1. Kang Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xuesong Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yuhu Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yuhu Cheng.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This work was supported by the National Natural Science Foundation of China under Grant 62176259 and Grant 61976215.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, K., Wang, X. & Cheng, Y. Few-shot learning based on enhanced pseudo-labels and graded pseudo-labeled data selection. Int. J. Mach. Learn. & Cyber. 14, 1783–1795 (2023). https://doi.org/10.1007/s13042-022-01727-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13042-022-01727-z

Keywords

Search

Navigation