双路径合作的原型矫正小样本分类模型

doi:10.3778/j.issn.1673-9418.2301070

摘要/Abstract

摘要： 基于度量的元学习在学习过程中存在由于稀缺数据分布导致习得的先验知识不足、从样本中提取到的单一视图特征易受弱相关或无关特征的干扰以及因分类造成的代表性特征偏差的问题。针对这些问题，提出了一种双路径合作的原型矫正小样本分类模型。首先，通过双路径合作模块从多视图角度自适应地突出关键特征和弱化弱相关特征，充分利用特征信息获得先验知识来提升特征的表达能力；其次，通过基于查询集样本特征信息的原型矫正分类策略来解决类内原型的偏差问题；最后，通过损失函数反向更新模型参数，模型分类准确率得以提升。在五个公开的数据集上进行了5-way 1-shot和5-way 5-shot对比实验，较基准模型而言，在miniImageNet数据集上，准确率提升了5.57个百分点和3.90个百分点；在tieredImageNet数据集上，准确率提升了5.68个百分点和3.93个百分点；在CUB数据集上，准确率提升了6.93个百分点和3.13个百分点；在CIFAR-FS数据集上，准确率提升了8.03个百分点和1.65个百分点；在FC-100数据集上，准确率提升了4.25个百分点和4.89个百分点。实验结果表明，提出的双路径合作的原型矫正小样本分类模型能在小样本学习领域有良好的性能，且模型中的模块可迁移到其他模型中使用。

关键词: 小样本学习, 元学习, 度量学习, 自适应双路径合作学习, 原型矫正

Abstract: In the learning process of the metric-based meta-learning, there are some problems, such as the lack of prior knowledge acquired due to the distribution of scarce data, the interference of weakly related or unrelated features extracted from a single-view sample, and the deviations of representative features caused by classification. To solve these problems, a prototype rectification few-shot classification model with dual-path cooperation is proposed in this paper. Firstly, the dual-path cooperation module adaptively highlights key features and weakens weakly related features from a multi-view perspective, and makes full use of feature information to obtain prior knowledge to improve the expression ability of features. Secondly, the problem of intra-class prototype with deviations is solved by the prototype rectification classification strategy with the sample feature information of the query set. Finally, the model parameters are updated reversely by means of the loss function, and the classification accuracy of the model is improved. Comparative experiments of 5-way 1-shot and 5-way 5-shot are conducted on five public datasets. Compared with baseline model, on the miniImageNet dataset, the accuracy is increased by 5.57 percentage points and 3.90 percentage points. On the tieredImageNet dataset, the accuracy is increased by 5.68 percentage points and 3.93 percentage points. On the CUB dataset, the accuracy is increased by 6.93 percentage points and 3.13 percentage points. On the CIFAR-FS dataset, the accuracy is increased by 8.03 percentage points and 1.65 percentage points. On the FC-100 dataset, the accuracy is increased by 4.25 percentage points and 4.89 percentage points. Experimental results show that the proposed model has good performance in the field of few-shot learning, and the modules in the model can be migrated to other models.

Key words: few-shot learning, meta-learning, metric learning, adaptive dual-path cooperation learning, prototype rectification

吕佳, 曾梦瑶, 董保森. 双路径合作的原型矫正小样本分类模型[J]. 计算机科学与探索, 2024, 18(3): 693-706.

LYU Jia, ZENG Mengyao, DONG Baosen. Prototype Rectification Few-Shot Classification Model with Dual-Path Cooperation[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(3): 693-706.

参考文献

[1] 汪航, 陈晓, 田晟兆, 等. 基于小样本学习的SAR图像识别[J]. 计算机科学, 2020, 47(5): 124-128.
WANG H, CHEN X, TIAN S Z, et al. SAR image recognition based on few-shot learning[J]. Computer Science, 2020, 47(5): 124-128.
[2] JANKOWSKI N, DUCH W, GR?BCZEWSKI K. Meta-learning in computational intelligence[M]. Berlin: Springer Science and Business Media, 2011: 97-115.
[3] LAKE B, SALAKHUTDINOV R. One-shot learning by inverting a compositional causal process[C]//Advances in Neural Information Processing Systems 26, Lake Tahoe, Dec 5-8, 2013: 2526-2534.
[4] THRUN S, PRATT L. Learning to learn: introduction and overview[M]//Learning to Learn. Boston: Springer-Verlag, 1998: 3-17.
[5] 李凡长, 刘洋, 吴鹏翔, 等. 元学习研究综述[J]. 计算机学报, 2021, 44(2): 422-446.
LI F C, LIU Y, WU P X, et al. A survey on recent advances in meta-learning[J]. Chinese Journal of Computers, 2021, 44(2): 422-446.
[6] KOCH G, ZEMEL R, SALAKHUTDINOV R. Siamese ne-ural networks for one-shot image recognition[C]//Proceedings of the 32nd International Conference on Machine Learning, Paris, Jul 10-11, 2015: 1-30.
[7] VINYALS O, BLUNDELL C, LILLICRAP T, et al. Matching networks for one shot learning[C]//Advances in Neural Information Processing Systems 29, Barcelona, Dec 5-10, 2016: 3630-3638.
[8] SNELL J, SWERSKY K, ZEMEL R S. Prototypical networks for few-shot learning[C]//Advances in Neural Information Processing Systems 30, Long Beach, Dec 4-9, 2017: 4077-4087.
[9] GAO T Y, HAN X, LIU Z, et al. Hybrid attention-based prototypical networks for noisy few-shot relation classification[C]//Proceedings of the 2019 AAAI Conference on Artificial Intelligence, Hawaii, Jan 27-Feb 1, 2019. Menlo Park: AAAI, 2019: 6407-6414.
[10] XU S L, ZHANG F, WEI X S, et al. Dual attention networks for few-shot fine-grained recognition[C]//Proceedings of the 2022 AAAI Conference on Artificial Intelligence, Dublin, May 22-27, 2022. Menlo Park: AAAI, 2022: 2911-2919.
[11] ZHOU B, KHOSLA A, LAPEDRIZA A, et al. Learning deep features for discriminative localization[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Jun 26-Jul 1, 2016. Washington: IEEE Computer Society, 2016: 2921-2929.
[12] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision, Munich, Sep 8-14, 2018. Cham: Springer, 2018: 3-19.
[13] REN M Y, TRIANTAFILLOU E, RAVI S, et al. Meta-learning for semi-supervised few-shot classification[J]. arXiv:1803. 00676, 2018.
[14] WAH C, BRANSON S, WELINDER P, et al. The Caltech-UCSD Birds-200-2011 dataset: CNS-TR-2011-001[R]. California Institute of Technology, 2011.
[15] BERTINETTO L, HENRIQUES J F, TORR P H S, et al. Meta-learning with differentiable closed-form solvers[J]. arXiv: 1805.08136, 2018.
[16] ORESHKIN B, RODRíGUEZ LóPEZ P, LACOSTE A. Tadam: task dependent adaptive metric for improved few-shot learning[C]//Advances in Neural Information Processing Systems 31, Montréal, Dec 3-8, 2018: 719-729.
[17] CAI Q, PAN Y, YAO T, et al. Memory matching networks for one-shot image recognition[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, Jun 18-21, 2018. Piscataway: IEEE, 2018: 4080-4088.
[18] KRIZHEVSKY A, HINTON G. Learning multiple layers of features from tiny images[D]. Toronto: University of Toronto, 2009.
[19] HE K, FAN H, WU Y, et al. Momentum contrast for unsupervised visual representation learning[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, Jun 14-19, 2020. Piscataway: IEEE, 2020: 9729-9738.
[20] LUO X, CHEN Y, WEN L, et al. Boosting few-shot classification with view-learnable contrastive learning[C]//Procee-dings of the 2021 IEEE International Conference on Multimedia and Expo, Shenzhen, Jul 5-9, 2021. Piscataway: IEEE, 2021: 1-6.
[21] LI A, LUO T, XIANG T, et al. Few-shot learning with global class representations[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Oct 27-Nov 2, 2019. Piscataway: IEEE, 2019: 9715-9724.
[22] LI W, WANG L, XU J, et al. Revisiting local descriptor based image-to-class measure for few-shot learning[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, Jun 16-20, 2019. Piscataway: IEEE, 2019: 7260-7268.
[23] HU Z, LI Z, WANG X, et al. Unsupervised descriptor selection based meta-learning networks for few-shot classification [J]. Pattern Recognition, 2022, 122: 108304.
[24] WANG Z, MA P, CHI Z, et al. Multi-attention mutual information distributed framework for few-shot learning[J]. Expert Systems with Applications, 2022, 202: 117062.
[25] MISHRA N, ROHANINEJAD M, CHEN X, et al. A simple neural attentive meta-learner[J]. arXiv:1707.03141, 2017.
[26] MUNKHDALAI T, YUAN X, MEHRI S, et al. Rapid adaptation with conditionally shifted neurons[C]//Proceedings of the 2018 International Conference on Machine Learning, Stockholm, May 26-28, 2018: 3664-3673.
[27] RAVICHANDRAN A, BHOTIKA R, SOATTO S. Few-shot learning with embedded class models and shot-free meta training[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Oct 27-Nov 2, 2019. Piscataway: IEEE, 2019: 331-339.
[28] LEE K, MAJI S, RAVICHANDRAN A, et al. Meta-learning with differentiable convex optimization[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, Jun 16-20, 2019. Piscataway: IEEE, 2019: 10657-10665.
[29] HOU R, CHANG H, MA B, et al. Cross attention network for few-shot classification[C]//Advances in Neural Information Processing Systems 32, Vancouver, Dec 8-14, 2019: 4005-4016.
[30] CHEN W Y, LIU Y C, KIRA Z, et al. A closer look at few-shot classification[C]//Proceedings of the 2019 International Conference on Learning Representations, New Orleans, May 6-9, 2019: 2-6.
[31] LIU Y, SCHIELE B, SUN Q. An ensemble of epoch-wise empirical Bayes for few-shot learning[C]//Proceedings of the 16th European Conference on Computer Vision, Aug 23-28, 2020. Cham: Springer, 2020: 404-421.
[32] GAO F, CAI L, YANG Z, et al. Multi-distance metric network for few-shot learning[J]. International Journal of Machine Learning and Cybernetics, 2022, 13(9): 2495-2506.
[33] GAO F, LUO X, YANG Z, et al. Label smoothing and task-adaptive loss function based on prototype network for few-shot learning[J]. Neural Networks, 2022, 156: 39-48.
[34] LI H, EIGEN D, DODGE S, et al. Finding task-relevant features for few-shot learning by category traversal[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, Jun 16-20, 2019. Piscataway: IEEE, 2019: 1-10.
[35] YANG F, WANG R, CHEN X. SEGA: semantic guided attention on visual prototype for few-shot learning[C]//Proceedings of the 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, Jun 5-9, 2022. Piscataway: IEEE, 2022: 1056-1066.
[36] FINN C, ABBEEL P, LEVINE S. Model-agnostic meta-learning for fast adaptation of deep networks[C]//Proceedings of the 2017 International Conference on Machine Learning, Sydney, Aug 6-11, 2017: 1126-1135.
[37] SUNG F, YANG Y, ZHANG L, et al. Learning to compare: relation network for few-shot learning[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, Jun 18-21, 2018. Piscataway: IEEE, 2018: 1199-1208.
[38] HUANG H, ZHANG J, YU L, et al. TOAN: target-oriented alignment network for fine-grained image categorization with few labeled samples[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 32(2): 853-866.
[39] YE H J, HU H, ZHAN D C, et al. Learning embedding adaptation for few-shot learning[J]. arXiv:1812.03664, 2018.
[40] ZHANG C, CAI Y, LIN G, et al. DeepEMD: few-shot image classification with differentiable earth mover’s distance and structured classifiers[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, Jun 14-19, 2020. Piscataway: IEEE, 2020: 12203 -12213.
[41] QIN Z, WANG H, MAWULI C B, et al. Multi-instance attention network for few-shot learning[J]. Information Sciences, 2022, 611: 464-475.
[42] HU S X, MORENO P G, XIAO Y, et al. Empirical Bayes transductive meta-learning with synthetic gradients[C]//Proceedings of the 2020 International Conference on Learning Representations, Addis Ababa, Apr 26-30, 2020.
[43] QIAO L, SHI Y, LI J, et al. Transductive episodic-wise adaptive metric for few-shot learning[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Oct 27-Nov 2, 2019. Piscataway: IEEE, 2019: 3603-3612.
[44] SIMON C, KONIUSZ P, NOCK R, et al. Adaptive subspaces for few-shot learning[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, Jun 14-19, 2020. Piscataway: IEEE, 2020: 4136- 4145.
[45] KIM J, KIM H, KIM G. Model-agnostic boundary-adversarial sampling for test-time generalization in few-shot learning[C]//Proceedings of the 16th European Conference on Computer Vision, Glasgow, Aug 23-28, 2020. Cham: Springer, 2020: 599-617.
[46] GIDARIS S, BURSUC A, KOMODAKIS N, et al. Boosting few-shot visual learning with self-supervision[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Oct 27-Nov 2, 2019. Piscataway: IEEE, 2019: 8059-8068.
[47] TIAN Y, WANG Y, KRISHNAN D, et al. Rethinking few-shot image classification: a good embedding is all you need?[C]//Proceedings of the 16th European Conference on Computer Vision, Glasgow, Aug 23-28, 2020. Cham: Springer, 2020: 266-282.
[48] OLIVER A, ODENA A, RAFFEL C A, et al. Realistic evaluation of deep semi-supervised learning algorithms[C]//Advances in Neural Information Processing Systems 31, Montreal, Dec 3-8, 2018: 3235-3246.
[49] WANG X, CAI J, JI S, et al. Self-adaptive label augmentation for semi-supervised few-shot classification[J]. arXiv: 2206.08150, 2022.
[50] LIU Y, LEE J, PARK M, et al. Learning to propagate labels: transductive propagation network for few-shot learning[J]. arXiv:1805.10002, 2018.