Abstract
This paper addresses the problem of machine learning for image classification, where a convolutional neural network (CNN) that has already learned the base classes learns novel classes from a small number of examples. Machine learning from a small number of examples is called few-shot learning (FSL). FSL that focuses on the classification performance of both base and novel classes in the joint space of these two classes is a challenging problem called generalized few-shot learning (GFSL). We propose a simple method for GFSL for image classification that adds the output unit of a novel class to the final layer of a CNN that has already trained on base classes and directly sets the weight vector of the output unit. Compared to existing methods, the proposed method has the advantage that the norms of the weight vectors of the final layer of the CNN need not be normalized during the training process on the base classes. We demonstrate the effectiveness of the proposed method on miniImageNet and tieredImageNet, two representative benchmark datasets for FSL.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Chen, Z., Fu, Y., Wang, Y.X., Ma, L., Liu, W., Hebert, M.: Image deformation meta-networks for one-shot learning. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8672–8681 (2019)
Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: Imagenet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255 (2009)
Dwivedi, S.K., Gupta, V., Mitra, R., Ahmed, S., Jain, A.: Protogan: towards few shot learning for action recognition. In: 2019 IEEE/CVF International Conference on Computer Vision Workshop, pp. 1308–1316 (2019)
Fan, Z., Ma, Y., Li, Z., Sun, J.: Generalized few-shot object detection without forgetting. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4525–4534 (2021)
Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep networks. In: Proceedings of the 34th International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 70, pp. 1126–1135 (2017)
Gidaris, S., Komodakis, N.: Dynamic few-shot visual learning without forgetting. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4367–4375 (2018)
Hariharan, B., Girshick, R.: Low-shot visual recognition by shrinking and hallucinating features. In: 2017 IEEE International Conference on Computer Vision, pp. 3037–3046 (2017)
Kukleva, A., Kuehne, H., Schiele, B.: Generalized and incremental few-shot learning by explicit learning and calibration without forgetting. In: 2021 IEEE/CVF International Conference on Computer Vision (2021)
Li, W., Wang, L., Xu, J., Huo, J., Gao, Y., Luo, J.: Revisiting local descriptor based image-to-class measure for few-shot learning. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7253–7260 (2019)
Li, Z., Zhou, F., Chen, F., Li, H.: Meta-SGD: learning to learn quickly for few-shot learning, arXiv preprint, arXiv:1707.09835 (2017)
Oreshkin, B., Rodríguez López, P., Lacoste, A.: Tadam: task dependent adaptive metric for improved few-shot learning. In: Advances in Neural Information Processing Systems, vol. 31, pp. 721–731 (2018)
Qi, H., Brown, M., Lowe, D.G.: Low-shot learning with imprinted weights. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5822–5830 (2018)
Ravi, S., Larochelle, H.: Optimization as a model for few-shot learning. In: International Conference on Learning Representations (2017)
Ren, M., Liao, R., Fetaya, E., Zemel, R.: Incremental few-shot learning with attention attractor networks. In: Advances in Neural Information Processing Systems, vol. 32 (2019)
Ren, M., et al.: Meta-learning for semi-supervised few-shot classification. In: International Conference on Learning Representations (2018)
Rusu, A.A., et al.: Meta-learning with latent embedding optimization. In: International Conference on Learning Representations (2019)
Schwartz, E., et al.: Delta-encoder: an effective sample synthesis method for few-shot object recognition. In: Advances in Neural Information Processing Systems, vol. 31 (2018)
Shi, X., Salewski, L., Schiegg, M., Welling, M.: Relational generalized few-shot learning. In: The 31st British Machine Vision Virtual Conference (2020)
Snell, J., Swersky, K., Zemel, R.: Prototypical networks for few-shot learning. In: Advances in Neural Information Processing Systems, vol. 30, pp. 4077–4087 (2017)
Sung, F., Yang, Y., Zhang, L., Xiang, T., Torr, P.H., Hospedales, T.M.: Learning to compare: relation network for few-shot learning. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1199–1208 (2018)
Vinyals, O., Blundell, C., Lillicrap, T., kavukcuoglu, k., Wierstra, D.: Matching networks for one shot learning. In: Advances in Neural Information Processing Systems, vol. 29, pp. 3630–3638 (2016)
Wang, Y.X., Girshick, R., Hebert, M., Hariharan, B.: Low-shot learning from imaginary data. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7278–7286 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Yamada, K. (2023). A Simple Method to Directly Determine Novel Class Weight of CNN for Generalized Few-Shot Learning. In: Lu, H., Blumenstein, M., Cho, SB., Liu, CL., Yagi, Y., Kamiya, T. (eds) Pattern Recognition. ACPR 2023. Lecture Notes in Computer Science, vol 14406. Springer, Cham. https://doi.org/10.1007/978-3-031-47634-1_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-47634-1_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-47633-4
Online ISBN: 978-3-031-47634-1
eBook Packages: Computer ScienceComputer Science (R0)