research-article

Learning to teach and learn for semi-supervised few-shot image classification

Authors:

Jianqiang Huang,

Qianru SunAuthors Info & Claims

Volume 212, Issue C

https://doi.org/10.1016/j.cviu.2021.103270

Published: 01 November 2021 Publication History

Abstract

This paper presents a novel semi-supervised few-shot image classification method named Learning to Teach and Learn (LTTL) to effectively leverage unlabeled samples in small-data regimes. Our method is based on self-training, which assigns pseudo labels to unlabeled data. However, the conventional pseudo-labeling operation heavily relies on the initial model trained by using a handful of labeled data and may produce many noisy labeled samples. We propose to solve the problem with three steps: firstly, cherry-picking searches valuable samples from pseudo-labeled data by using a soft weighting network; and then, cross-teaching allows the classifiers to teach mutually for rejecting more noisy labels. A feature synthesizing strategy is introduced for cross-teaching to avoid clean samples being rejected by mistake; finally, the classifiers are fine-tuned with a few labeled data to avoid gradient drifts. We use the meta-learning paradigm to optimize the parameters in the whole framework. The proposed LTTL combines the power of meta-learning and self-training, achieving superior performance compared with the baseline methods on two public benchmarks.

Highlights

•

We propose a novel self-training strategy for semi-supervised few-shot image classification.

•

We propose a meta-learned cherry-picking operation for selecting valuable training samples.

•

A cross-teaching paradigm is designed for the inner-loop updating process of meta-learning, enabling the classifiers to learn from noisy-labeled data.

References

[1]

Antoniou, A., Edwards, H., Storkey, A., 2019. How to train your maml. In: ICLR.

[2]

Arazo, E., Ortego, D., Albert, P., O’Connor, N.E., McGuinness, K., 2019. Unsupervised label noise modeling and loss correction. In: ICML.

[3]

Arpit, D., Jastrzebski, S., Ballas, N., Krueger, D., Bengio, E., Kanwal, M.S., Maharaj, T., Fischer, A., Courville, A.C., Bengio, Y., Lacoste-Julien, S., 2017. A closer look at memorization in deep networks. In: ICML.

[4]

Berthelot, D., Carlini, N., Goodfellow, I.J., Papernot, N., Oliver, A., Raffel, C., 2019. MixMatch: A holistic approach to semi-supervised learning. In: NeurIPS.

[5]

Chen, W.-Y., Liu, Y.-C., Kira, Z., Wang, Y.-C., Huang, J.-B., 2019. A closer look at few-shot classification. In: ICLR.

[6]

Dong-Hyun, L., 2013. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In: ICML Workshops.

[7]

Finn, C., Abbeel, P., Levine, S., 2017. Model-agnostic meta-learning for fast adaptation of deep networks. In: ICML.

[8]

Finn, C., Xu, K., Levine, S., 2018. Probabilistic model-agnostic meta-learning. In: NeurIPS.

[9]

Franceschi, L., Frasconi, P., Salzo, S., Grazzi, R., Pontil, M., 2018. Bilevel programming for hyperparameter optimization and meta-learning. In: ICML.

[10]

Grant, E., Finn, C., Levine, S., Darrell, T., Griffiths, T.L., 2018. Recasting gradient-based meta-learning as hierarchical Bayes. In: ICLR.

[11]

Han, B., Yao, Q., Yu, X., Niu, G., Xu, M., Hu, W., Tsang, I.W., Sugiyama, M., 2018. Co-teaching: Robust training of deep neural networks with extremely noisy labels. In: NeurIPS.

[12]

He, K., Zhang, X., Ren, S., Sun, J., 2016. Deep residual learning for image recognition. In: CVPR.

[13]

Hou, R., Chang, H., Bingpeng, M., Shan, S., Chen, X., 2019. Cross attention network for few-shot classification. In: NeurIPS.

[14]

Huang, J., Qu, L., Jia, R., Zhao, B., 2019. O2U-net: A simple noisy label detection approach for deep neural networks. In: ICCV.

[15]

Laine, S., Aila, T., 2017. Temporal ensembling for semi-supervised learning. In: ICLR.

[16]

Li, J., Socher, R., Hoi, S.C., 2020. DivideMix: Learning with noisy labels as semi-supervised learning. In: ICLR.

[17]

Li, X., Sun, Q., Liu, Y., Zheng, S., Zhou, Q., Chua, T.-S., Schiele, B., 2019. Learning to self-train for semi-supervised few-shot classification. In: NeurIPS.

[18]

Liu, Y., Lee, J., Park, M., Kim, S., Yang, Y., 2019. Transductive propagation network for few-shot learning. In: ICLR.

[19]

Liu, Y., Schiele, B., Sun, Q., 2020. An Ensemble of Epoch-Wise Empirical Bayes for Few-Shot Learning. In: ECCV.

[20]

Miyato T., Dai A.M., Goodfellow I.J., Virtual adversarial training for semi-supervised text classification, 2016, ArXiv 1605.07725.

[21]

Ravi, S., Larochelle, H., 2017. Optimization as a model for few-shot learning. In: ICLR.

[22]

Ren, M., Triantafillou, E., Ravi, S., Snell, J., Swersky, K., Tenenbaum, J.B., Larochelle, H., Zemel, R.S., 2018. Meta-learning for semi-supervised few-shot classification. In: ICLR.

[23]

Ren, M., Zeng, W., Yang, B., Urtasun, R., 2018. Learning to reweight examples for robust deep learning. In: ICML.

[24]

Rohrbach, M., Ebert, S., Schiele, B., 2013. Transfer learning in a transductive setting. In: NIPS.

[25]

Russakovsky O., Deng J., Su H., Krause J., Satheesh S., Ma S., Huang Z., Karpathy A., Khosla A., Bernstein M., Berg A.C., Fei-Fei L., ImageNet Large scale visual recognition challenge, Int. J. Comput. Vis. 115 (3) (2015) 211–252.

Digital Library

[26]

Rusu, A.A., Rao, D., Sygnowski, J., Vinyals, O., Pascanu, R., Osindero, S., Hadsell, R., 2019. Meta-learning with latent embedding optimization. In: ICLR.

[27]

Shelhamer E., Long J., Darrell T., Fully convolutional networks for semantic segmentation, IEEE Trans. Pattern Anal. Mach. Intell. 39 (4) (2017) 640–651.

Digital Library

[28]

Simon, C., Koniusz, P., Nock, R., Harandi, M., 2020. Adaptive subspaces for few-shot learning. In: CVPR.

[29]

Snell, J., Swersky, K., Zemel, R.S., 2017. Prototypical networks for few-shot learning. In: NIPS.

[30]

Sun Q., Liu Y., Chen Z., Chua T.-S., Schiele B., Meta-transfer learning through hard tasks, 2019, ArXiv 1910.03648.

[31]

Sun, Q., Liu, Y., Chua, T.-S., Schiele, B., 2019b. Meta-transfer learning for few-shot learning. In: CVPR.

[32]

Sung, F., Yang, Y., Zhang, L., Xiang, T., Torr, P.H.S., Hospedales, T.M., 2018. Learning to compare: relation network for few-shot learning. In: CVPR.

[33]

Triguero I., García S., Herrera F., Self-labeled techniques for semi-supervised learning: taxonomy, software and empirical study, Knowl. Inf. Syst. 42 (2) (2015) 245–284.

[34]

Vinyals, O., Blundell, C., Lillicrap, T., Kavukcuoglu, K., Wierstra, D., 2016. Matching networks for one shot learning. In: NIPS.

[35]

Wang, Y., Girshick, R.B., Hebert, M., Hariharan, B., 2018. Low-shot learning from imaginary data. In: CVPR.

[36]

Wang, X., Wang, S., Shi, H., Wang, J., Mei, T., 2019. Co-mining: Deep face recognition with noisy labels. In: ICCV.

[37]

Xian, Y., Sharma, S., Schiele, B., Akata, Z., 2019. f-VAEGAN-D2: A feature generating framework for any-shot learning. In: CVPR.

[38]

Yann L., Yoshua B., Geoffrey H., Deep learning, Nature 521 (7553) (2015) 436.

[39]

Yarowsky, D., 1995. Unsupervised word sense disambiguation rivaling supervised methods. In: ACL.

[40]

Yu, Z., Chen, L., Cheng, Z., Luo, J., 2020. TransMatch: A transfer-learning scheme for semi-supervised few-shot learning. In: CVPR.

[41]

Yu, X., Han, B., Yao, J., Niu, G., Tsang, I.W., Sugiyama, M., 2019. How does disagreement help generalization against label corruption? In: ICML.

[42]

Zhang, R., Che, T., Grahahramani, Z., Bengio, Y., Song, Y., 2018b. MetaGAN: An adversarial approach to few-shot learning. In: NeurIPS.

[43]

Zhang, H., Cissé, M., Dauphin, Y.N., Lopez-Paz, D., 2018a. mixup: Beyond empirical risk minimization. In: ICLR.

Cited By

Mahalakshmi VSandhu MShabaz MKeshta IPrasad KKuzieva NByeon HSoni M(2024)Few-shot learning-based human behavior recognition modelComputers in Human Behavior10.1016/j.chb.2023.108038151:COnline publication date: 4-Mar-2024
https://dl.acm.org/doi/10.1016/j.chb.2023.108038

Index Terms

Learning to teach and learn for semi-supervised few-shot image classification
1. Computing methodologies
  1. Artificial intelligence
    1. Computer vision
      1. Computer vision problems
  2. Machine learning
    1. Learning paradigms
      1. Supervised learning
        Supervised learning by classification
    2. Machine learning approaches
      1. Classification and regression trees

Index terms have been assigned to the content through auto-classification.

Recommendations

Improved prototypical network for active few-shot learning
Highlights
- The novel active few-shot learning task is proposed to better accomplish classification with few labeled examples.
- AC-FSL model is designed by introducing the active loss prediction module into the few-shot prototype network.
- ...
Abstract
Different from deep learning with large scale supervision, few-shot learning aims to learn the samples’ characteristics from few labeled examples. Apparently, few-shot learning is more in line with the visual cognitive mechanism of the human ...
Weakly supervised label distribution learning based on transductive matrix completion with sample correlations
Highlights
- The matrix completion is firstly applied to weakly supervised LDL problems.
- ...
Abstract
Label distribution learning (LDL) is one of the paradigms for dealing with label ambiguity, and it can learn the relative importance of each label to a particular instance. Most of the existing LDL approaches require strong supervision ...
Global-guided weakly-supervised learning for multi-label image classification
Abstract
Multi-label classification with region-free labels is attracting increasing attention compared to that with region-based labels due to the time-consuming manual region-labeling process. Existing methods usually employ attention-based technology ...
Highlights
- The proposed methods designs a global-guided feature disentanglement architecture to improve the feature representation ability.
- The global correlations of multi-level features are used to guide the feature disentanglement of each ...

Comments

Information & Contributors

Information

Published In

cover image Computer Vision and Image Understanding

Computer Vision and Image Understanding Volume 212, Issue C

Nov 2021

114 pages

ISSN:1077-3142

Issue’s Table of Contents

Elsevier Inc.

Publisher

Elsevier Science Inc.

United States

Publication History

Published: 01 November 2021

Author Tags

Author Tags

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

1
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 23 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Mahalakshmi VSandhu MShabaz MKeshta IPrasad KKuzieva NByeon HSoni M(2024)Few-shot learning-based human behavior recognition modelComputers in Human Behavior10.1016/j.chb.2023.108038151:COnline publication date: 4-Mar-2024
https://dl.acm.org/doi/10.1016/j.chb.2023.108038

View Options

View options

Media

Figures

Other

Tables

View Issue’s Table of Contents