Article

Rethinking Few-Shot Image Classification: A Good Embedding is All You Need?

Authors:

Dilip Krishnan,

Joshua B. Tenenbaum,

Phillip IsolaAuthors Info & Claims

Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XIV

Pages 266 - 282

https://doi.org/10.1007/978-3-030-58568-6_16

Published: 23 August 2020 Publication History

Abstract

The focus of recent meta-learning research has been on the development of learning algorithms that can quickly adapt to test time tasks with limited data and low computational cost. Few-shot learning is widely used as one of the standard benchmarks in meta-learning. In this work, we show that a simple baseline: learning a supervised or self-supervised representation on the meta-training set, followed by training a linear classifier on top of this representation, outperforms state-of-the-art few-shot learning methods. An additional boost can be achieved through the use of self-distillation. This demonstrates that using a good learned embedding model can be more effective than sophisticated meta-learning algorithms. We believe that our findings motivate a rethinking of few-shot image classification benchmarks and the associated role of meta-learning algorithms. Code: http://github.com/WangYueFt/rfs/.

References

[1]

Machine learning in python. https://scikit-learn.org/stable/

[2]

Allen, K., Shelhamer, E., Shin, H., Tenenbaum, J.: Infinite mixture prototypes for few-shot learning. In: ICML (2019)

[3]

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

[4]

Buciluǎ, C., Caruana, R., Niculescu-Mizil, A.: Model compression. In: SIGKDD (2006)

[5]

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

[6]

Chen, Y., Wang, X., Liu, Z., Xu, H., Darrell, T.: A new meta-baseline for few-shot learning. ArXiv abs/2003.04390 (2020)

[7]

Clark, K., Luong, M.T., Manning, C.D., Le, Q.V.: Bam! born-again multi-task networks for natural language understanding. In: ACL (2019)

[8]

Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: CVPR09 (2009)

[9]

Dhillon, G.S., Chaudhari, P., Ravichandran, A., Soatto, S.: A baseline for few-shot image classification. In: ICLR (2020)

[10]

Dvornik, N., Schmid, C., Mairal, J.: Diversity with cooperation: ensemble methods for few-shot classification. In: ICCV (2019)

[11]

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

[12]

Furlanello, T., Lipton, Z.C., Tschannen, M., Itti, L., Anandkumar, A.: Born-again neural networks. In: ICML (2018)

[13]

Gan, C., Gong, B., Liu, K., Su, H., Guibas, L.J.: Geometry guided convolutional neural networks for self-supervised video representation learning. In: CVPR (2018)

[14]

Gan, C., Zhao, H., Chen, P., Cox, D., Torralba, A.: Self-supervised moving vehicle tracking with stereo sound. In: ICCV (2019)

[15]

Gidaris, S., Komodakis, N.: Dynamic few-shot visual learning without forgetting. In: CVPR (2018)

[16]

Hao, F., He, F., Cheng, J., Wang, L., Cao, J., Tao, D.: Collect and select: semantic alignment metric learning for few-shot learning. In: ICCV (2019)

[17]

He, K., Fan, H., Wu, Y., Xie, S., Girshick, R.B.: Momentum contrast for unsupervised visual representation learning. ArXiv abs/1911.05722 (2019)

[18]

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

[19]

Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. In: NIPS Deep Learning and Representation Learning Workshop (2015)

[20]

Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: CVPR (2018)

[21]

Huang, S., Tao, D.: All you need is a good representation: A multi-level and classifier-centric representation for few-shot learning. ArXiv abs/1911.12476 (2019)

[22]

Jamal, M.A., Qi, G.J.: Task agnostic meta-learning for few-shot learning. In: CVPR (2019)

[23]

Koch, G., Zemel, R., Salakhutdinov, R.: Siamese neural networks for one-shot image recognition. In: ICML Deep Learning Workshop (2015)

[24]

Lake BM, Salakhutdinov R, and Tenenbaum JB Human-level concept learning through probabilistic program induction Science 2015 350 6266 1332-1338

[25]

Lake BM, Salakhutdinov R, and Tenenbaum JB The Omniglot challenge: a 3-year progress report Curr. Opin. Behav. Sci. 2019 29 97-104

[26]

Lee, K., Maji, S., Ravichandran, A., Soatto, S.: Meta-learning with differentiable convex optimization. In: CVPR (2019)

[27]

Li, A., Luo, T., Xiang, T., Huang, W., Wang, L.: Few-shot learning with global class representations. In: ICCV (2019)

[28]

Li, H., Eigen, D., Dodge, S., Zeiler, M., Wang, X.: Finding task-relevant features for few-shot learning by category traversal. In: CVPR (2019)

[29]

Mishra, N., Rohaninejad, M., Chen, X., Abbeel, P.: A simple neural attentive meta-learner. arXiv preprint arXiv:1707.03141 (2017)

[30]

Mobahi, H., Farajtabar, M., Bartlett, P.L.: Self-distillation amplifies regularization in hilbert space. arXiv preprint arXiv:2002.05715 (2020)

[31]

Munkhdalai, T., Yuan, X., Mehri, S., Trischler, A.: Rapid adaptation with conditionally shifted neurons. arXiv preprint arXiv:1712.09926 (2017)

[32]

Nichol, A., Achiam, J., Schulman, J.: On first-order meta-learning algorithms. ArXiv abs/1803.02999 (2018)

[33]

Oreshkin, B., López, P.R., Lacoste, A.: Tadam: task dependent adaptive metric for improved few-shot learning. In: NIPS (2018)

[34]

Peng, Z., Li, Z., Zhang, J., Li, Y., Qi, G.J., Tang, J.: Few-shot image recognition with knowledge transfer. In: ICCV (2019)

[35]

Qiao, L., Shi, Y., Li, J., Wang, Y., Huang, T., Tian, Y.: Transductive episodic-wise adaptive metric for few-shot learning. In: ICCV (2019)

[36]

Qiao, S., Liu, C., Shen, W., Yuille, A.L.: Few-shot image recognition by predicting parameters from activations. In: CVPR (2018)

[37]

Raghu, A., Raghu, M., Bengio, S., Vinyals, O.: Rapid learning or feature reuse? towards understanding the effectiveness of maml. arXiv preprint arXiv:1909.09157 (2019)

[38]

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

[39]

Ravichandran, A., Bhotika, R., Soatto, S.: Few-shot learning with embedded class models and shot-free meta training. In: ICCV (2019)

[40]

Ren, M., et al.: Meta-learning for semi-supervised few-shot classification. In: ICLR (2018)

[41]

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

[42]

Scott, T., Ridgeway, K., Mozer, M.C.: Adapted deep embeddings: a synthesis of methods for k-shot inductive transfer learning. In: NIPS (2018)

[43]

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

[44]

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

[45]

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

[46]

Tian, Y., Krishnan, D., Isola, P.: Contrastive multiview coding. arXiv preprint arXiv:1906.05849 (2019)

[47]

Tian, Y., Krishnan, D., Isola, P.: Contrastive representation distillation. arXiv preprint arXiv:1910.10699 (2019)

[48]

Tian, Y., Sun, C., Poole, B., Krishnan, D., Schmid, C., Isola, P.: What makes for good views for contrastive learning? arXiv preprint arXiv:2005.10243 (2020)

[49]

Triantafillou, E., Zemel, R.S., Urtasun, R.: Few-shot learning through an information retrieval lens. In: NIPS (2017)

[50]

Triantafillou, E., et al.: Meta-dataset: a dataset of datasets for learning to learn from few examples. arXiv preprint arXiv:1903.03096 (2019)

[51]

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

[52]

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

[53]

Wang YX and Hebert M Learning from small sample sets by combining unsupervised meta-training with CNNs Adv. Neural Inform. Process. Syst. 2016 29 244-252

[54]

Wang, Y., Hebert, M.: Learning to learn: model regression networks for easy small sample learning. In: ECCV (2016)

[55]

Weng, L.: Meta-learning: Learning to learn fast. lilianweng.github.io/lil-log (2018). http://lilianweng.github.io/lil-log/2018/11/29/meta-learning.html

[56]

Wu, Z., Xiong, Y., Yu, S.X., Lin, D.: Unsupervised feature learning via non-parametric instance discrimination. In: CVPR (2018)

[57]

Wu, Z., Li, Y., Guo, L., Jia, K.: Parn: position-aware relation networks for few-shot learning. In: ICCV (2019)

[58]

Ye, H.J., Hu, H., Zhan, D.C., Sha, F.: Learning embedding adaptation for few-shot learning. CoRR abs/1812.03664 (2018)

[59]

Yim, J., Joo, D., Bae, J., Kim, J.: A gift from knowledge distillation: fast optimization, network minimization and transfer learning. In: CVPR (2017)

[60]

Zhang, J., Zhao, C., Ni, B., Xu, M., Yang, X.: Variational few-shot learning. In: ICCV (2019)

Cited By

Bi YZhu HShi JSong B(2025)TsCANet: Three-stream contrastive adaptive network for cross-domain few-shot learningThe Journal of Supercomputing10.1007/s11227-024-06482-281:1Online publication date: 1-Jan-2025
https://dl.acm.org/doi/10.1007/s11227-024-06482-2
Yang XYao HWei YSalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)One meta-tuned transformer is what you need for few-shot learningProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694409(56681-56703)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3694409
Tian HLiu FLiu TDu BCheung YHan BSalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)MOKDProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694038(48154-48185)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3694038
Show More Cited By

Recommendations

Improving Few-Shot Image Classification with Self-supervised Learning
Cloud Computing – CLOUD 2022
Abstract
Few-Shot Image Classification (FSIC) aims to learn an image classifier with only a few training samples. The key challenge of few-shot image classification is to learn this classifier with scarce labeled data. To tackle the issue, we leverage the ...
Self-Supervision Can Be a Good Few-Shot Learner
Computer Vision – ECCV 2022
Abstract
Existing few-shot learning (FSL) methods rely on training with a large labeled dataset, which prevents them from leveraging abundant unlabeled data. From an information-theoretic perspective, we propose an effective unsupervised FSL method, ...
Dual class representation learning for few-shot image classification
Abstract
Few-shot learning (FSL) models are trained on base classes that have many training examples and evaluated on novel classes that have very few training examples. Since these models cannot be properly fine-tuned on the novel classes ...
Highlights
- Proposes dual class representation learning (DCRL) for few-shot image classification.

Comments

Information & Contributors

Information

Published In

cover image Guide Proceedings

Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XIV

Aug 2020

842 pages

ISBN:978-3-030-58567-9

DOI:10.1007/978-3-030-58568-6

Editors:
Andrea Vedaldi
University of Oxford, Oxford, UK
,
Horst Bischof
Graz University of Technology, Graz, Austria
,
Thomas Brox
University of Freiburg, Freiburg im Breisgau, Germany
,
Jan-Michael Frahm
University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

© Springer Nature Switzerland AG 2020.

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 23 August 2020

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

90
Total Citations
View Citations
0
Total Downloads

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

Reflects downloads up to 04 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Bi YZhu HShi JSong B(2025)TsCANet: Three-stream contrastive adaptive network for cross-domain few-shot learningThe Journal of Supercomputing10.1007/s11227-024-06482-281:1Online publication date: 1-Jan-2025
https://dl.acm.org/doi/10.1007/s11227-024-06482-2
Yang XYao HWei YSalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)One meta-tuned transformer is what you need for few-shot learningProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694409(56681-56703)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3694409
Tian HLiu FLiu TDu BCheung YHan BSalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)MOKDProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694038(48154-48185)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3694038
Wang JLiu XSu Y(2024)A Robust Few-shot Learning Framework via Dual-branch Adversarial Noise PretrainingProceedings of the 6th ACM International Conference on Multimedia in Asia10.1145/3696409.3700245(1-8)Online publication date: 3-Dec-2024
https://dl.acm.org/doi/10.1145/3696409.3700245
Zhou XLiu FZhang CLi FCai WZhou J(2024)Feature-weighted Multi-stage Bayesian Prototype for Few-shot ClassificationProceedings of the 6th ACM International Conference on Multimedia in Asia10.1145/3696409.3700244(1-7)Online publication date: 3-Dec-2024
https://dl.acm.org/doi/10.1145/3696409.3700244
Gong NDuan PRong Y(2024)Learning Task-Specific Embeddings for Few-Shot Classification via Local Weight AdaptationProceedings of the 2024 16th International Conference on Machine Learning and Computing10.1145/3651671.3651746(485-491)Online publication date: 2-Feb-2024
https://dl.acm.org/doi/10.1145/3651671.3651746
Zhan QWang BJiang AXie XZhang MLiu G(2024)A two-stage spiking meta-learning method for few-shot classificationKnowledge-Based Systems10.1016/j.knosys.2023.111220284:COnline publication date: 25-Jan-2024
https://dl.acm.org/doi/10.1016/j.knosys.2023.111220
Cao ZGuo FAn YWang J(2024)Residual Spatio-Temporal Attention Based Prototypical Network for Rare Arrhythmia ClassificationBioinformatics Research and Applications10.1007/978-981-97-5087-0_8(89-101)Online publication date: 19-Jul-2024
https://dl.acm.org/doi/10.1007/978-981-97-5087-0_8
Paul RVora SThakur NLi B(2024)A-FSL: Adaptive Few-Shot Learning via Task-Driven Context Aggregation and Attentive Feature RefinementPattern Recognition10.1007/978-3-031-78395-1_7(97-113)Online publication date: 1-Dec-2024
https://dl.acm.org/doi/10.1007/978-3-031-78395-1_7
Chen LHe ZZhang H(2024)Image Domain Translation for Few-Shot LearningPattern Recognition10.1007/978-3-031-78183-4_20(313-329)Online publication date: 1-Dec-2024
https://dl.acm.org/doi/10.1007/978-3-031-78183-4_20
Show More Cited By

View Options

View options

Media

Figures

Other

Tables

View Table of Contents