Abstract
For real-world graph data, the node class distribution is inherently imbalanced and long-tailed, which naturally leads to a few-shot learning scenario with limited nodes labeled for newly emerging classes. There are many carefully designed solutions for such a few-shot learning problem via methods such as data augmentation, learning transferable initialization, learning prototypes, and many more. However, most, if not all, of them are based on a strong assumption that all the test nodes exclusively come from novel classes, which is impractical in real-world applications. In this paper, we study a broader and more realistic problem named generalized few-shot node classification, where the test samples can be from both novel classes and base classes. Compared with the standard few-shot node classification, this new problem imposes several unique challenges, including asymmetric classification and inconsistent preference. To counter those challenges, we propose a shot-aware neural node classifier (Stager) equipped with an uncertainty-based weight assigner module for adaptive propagation. As the existing meta-learning solutions cannot handle this new problem, we propose a novel training paradigm named imbalanced episodic training to ensure the label distribution is consistent between the meta-training and meta-test scenarios. Comprehensive experiments on four real-world datasets demonstrate the effectiveness of our proposed model and training paradigm.
Similar content being viewed by others
Notes
References
Aggarwal CC, Nan L (2011) On node classification in dynamic content-based networks. In: SDM
Alaa A, Van Der Schaar M (2020) Discriminative jackknife: quantifying uncertainty in deep learning via higher-order influence functions. In: ICML
Smriti B, Graham C, Muthukrishnan S (2011) Node classification in social networks. Social network data analytics. Springer, New York, pp 115–148
Bojchevski A, Günnemann S (2018) Deep gaussian embedding of graphs: unsupervised inductive learning via ranking. In: ICLR
Bose AJ, Jain A, Molino P, and Hamilton WL (2019) Meta-graph: few shot link prediction via meta learning. arXiv preprint arXiv:1912.09867
Buda M, Maki A, Mazurowski MA (2018) A systematic study of the class imbalance problem in convolutional neural networks. Neural Netw 106:249–259
Chao W-L, Changpinyo S, Gong B, and Sha F (2016) An empirical study and analysis of generalized zero-shot learning for object recognition in the wild. In: ECCV
Chauhan J, Nathani D, and Kaul M (2020) Few-shot learning on graphs via super-classes based on graph spectral measures. In: ICLR
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) SMOTE: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357
Chen X, Dai H, Li Y, Gao X, and Song L (2020) Learning to stop while learning to predict. In: International conference on machine learning. PMLR, pp 1520–1530
Chien E, Peng J, Li P, and Milenkovic O (2021) Adaptive universal generalized pagerank graph neural network. In: ICLR
Ding K, Wang J, Caverlee J, and Liu H (2022) Meta propagation networks for graph few-shot semi-supervised learning. In: AAAI
Ding K, Wang J, Li J, Shu K, Liu C, and Liu H (2020) Graph prototypical networks for few-shot learning on attributed networks. In: CIKM
Ding K, Xu Z, Tong H, and Liu H (2022) Data augmentation for deep graph learning: a survey. arXiv preprint arXiv:2202.08235
Ding K, Zhou Q, Tong H, and Liu H (2021) Few-shot network anomaly detection via cross-network meta-learning. In: TheWebConf
Elkan C (2001) The foundations of cost-sensitive learning. In: IJCAI
Fan Z, Ma Y, Li Z, and Sun J (2021) Generalized few-shot object detection without forgetting. In: CVPR
Felix R, Reid I, Carneiro G, et al (2018) Multi-modal cycle-consistent generalized zero-shot learning. In: ECCV
Finn C, Abbeel P, and Levine S (2017) Model-agnostic meta-learning for fast adaptation of deep networks. In: ICML
Franceschi L, Donini M, Frasconi P, Pontil M (2017) Forward and reverse gradient-based hyperparameter optimization. In: ICML. PMLR
Franceschi L, Frasconi P, Salzo S, Grazzi R, Pontil M (2018) Bilevel programming for hyperparameter optimization and meta-learning. In: ICML. PMLR
Gal Y, and Ghahramani Z (2015) Bayesian convolutional neural networks with bernoulli approximate variational inference. arXiv preprint arXiv:1506.02158
Gidaris S, and Komodakis N (2018) Dynamic few-shot visual learning without forgetting. In: CVPR
Guan J, Liu J, Sun J, Feng P, Shuai T, and Wang W (2020) Meta metric learning for highly imbalanced aerial scene classification. In: ICASSP 2020-2020 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, pp 4047–4051
He H, and Garcia EA (2009) Learning from imbalanced data. TKDE
Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2(5):359–366
Kexin H, Marinka Z (2020) Graph meta learning via local subgraphs. NeurIPS 33:5862–5874
Japkowicz N, Stephen S (2002) The class imbalance problem: a systematic study. Intell Data Anal 6(5):429–449
Kendall A, and Gal Y (2017) What uncertainties do we need in bayesian deep learning for computer vision? In: NIPS
Kingma DP, and Ba J (2015) Adam: a method for stochastic optimization. In: ICLR
Thomas N (2017) Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In: ICLR
Klicpera J, Bojchevski A, and Stephan G. (2018) Graph neural networks meet personalized pagerank. In: ICLR, Predict then propagate
Lan L, Wang P, Du X, Song K, Tao J, Guan X (2020) Node classification on graphs with few-shot novel labels via meta transformed network embedding. Adv Neural Inf Process Syst 33:16520–16531
Lee H, Lee H, Na D, Kim S, Park M, Yang E, and Hwang SJ (2020) Learning to balance: bayesian meta-learning for imbalanced and out-of-distribution tasks. In: 8th international conference on learning representations, ICLR 2020, Addis Ababa, Ethiopia, April 26–30, 2020. OpenReview.net
Liu Z, Fang Y, Liu C, and Hoi SCH (2021) Relative and absolute location embedding for few-shot node classification on graph. In: AAAI
Liu Z, Miao Z, Zhan X, Wang J, Gong B, and Yu SX (2019) Large-scale long-tailed recognition in an open world. In: CVPR
McAuley J, Pandey R, and Leskovec J (2015) Inferring networks of substitutable and complementary products. In: SIGKDD
Nichol A, Achiam J, and Schulman J (2018) On first-order meta-learning algorithms. arXiv preprint arXiv:1803.02999
Perez-Rua J-M, Zhu X, Timothy MH, and Tao X (2020) Incremental few-shot object detection. In: CVPR
Qu L, Zhu H, Zheng R, Shi Y, and Yin H (2021) Imgagn: imbalanced network embedding via generative adversarial graph networks. In: Feida Z, Beng CO, and Chunyan M (eds) SIGKDD
Snell J, Swersky K, and Zemel R (2017) Prototypical networks for few-shot learning. In: NIPS
Ni Srivastava, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Tan J, Wang C, Li B, Li Q, Ouyang W, Yin C, and Yan J (2020) Equalization loss for long-tailed object recognition. In: CVPR
Tang J, Zhang J, Yao L, Li J, Zhang L, and Su Z (2008) Arnetminer: extraction and mining of academic social networks. In: SIGKDD
Tang J, Aggarwal C, and Liu H (2016) Node classification in signed social networks. In: SDM
Thrun S (1996) Is learning the n-th thing any easier than learning the first? In: NIPS
Thrun S (1998) Lifelong learning algorithms. Learning to learn. Springer, New York, pp 181–209
Ting K M (2002) An instance-weighting method to induce cost-sensitive trees. TKDE
Triantafillou E, Zhu T, Dumoulin V, Lamblin P, Evci U, Xu K, Goroshin R, Gelada C, Swersky K, Manzagol P-A, and Larochelle H (2020) Meta-dataset: a dataset of datasets for learning to learn from few examples. In: 8th international conference on learning representations. ICLR 2020, Addis Ababa, Ethiopia, April 26–30, 2020. OpenReview.net
Petar V, Guillem C, Arantxa C, Adriana R, Pietro L, and Yoshua B (2018) Graph attention networks. In: ICLR
Vinyals O, Blundell C, Lillicrap T, Wierstra D, et al (2016) Matching networks for one shot learning. In: NIPS
Wang N, Luo M, Ding K, Zhang L, Li J, and Zheng Q (2020) Graph few-shot learning with attribute matching. In: CIKM
Wang Y-X, Ross G, Martial H, and Bharath H (2018) Low-shot learning from imaginary data. In: CVPR
Gary MW, Foster P (2003) Learning when training data are costly: the effect of class distribution on tree induction. J Artif Intell Res 19:315–354
Felix W, Souza A, Tianyi Z, Tao Y, and Kilian W (2019) Simplifying graph convolutional networks. In: ICML. Christopher Fifty
Zhe X, Boxin D, Tong H (2022) Graph sanitation with application to node classification. In: Proceedings of the ACM web conference. pp 1136–1147
Yao H, Zhang C, Wei Y, Jiang M, Wang S, Huang J, Chawla N, and Li Z (2020) Graph few-shot learning via knowledge transfer. In: AAAI
Ye H-J, Hu H, and Zhan D-C (2021) Learning adaptive classifiers synthesis for generalized few-shot learning. Int J Comput Vis 1–24
Yu Y, Ji Z, Han J, and Zhang Z (2020) Episode-based prototype generating network for zero-shot learning. In: CVPR
Zhao T, Zhang X, and Suhang W (2021) Imbalanced node classification on graphs with graph neural networks. In: WSDM. Graphsmote
Zhou F, Cao C, Zhang K, Trajcevski G, Zhong T and Ji G (2019) On few-shot node classification in graph meta-learning. In: CIKM. Meta-gnn
Author information
Authors and Affiliations
Contributions
All authors discussed the methodology and algorithm. Zhe Xu and Dr. Hanghang Tong wrote the main manuscript. Zhe did the experiments.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
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.
About this article
Cite this article
Xu, Z., Ding, K., Wang, YX. et al. Generalized few-shot node classification: toward an uncertainty-based solution. Knowl Inf Syst 66, 1205–1229 (2024). https://doi.org/10.1007/s10115-023-01975-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10115-023-01975-7