Geometer: Graph Few-Shot Class-Incremental Learning via Prototype Representation

被引:15
作者
Lu, Bin [1 ]
Gan, Xiaoying [1 ]
Yang, Lina [1 ]
Zhang, Weinan [1 ]
Fu, Luoyi [1 ]
Wang, Xinbing [1 ]
机构
[1] Shanghai Jiao Tong Univ, Shanghai, Peoples R China
来源
PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2022 | 2022年
关键词
Node Classification; Graph Neural Network; Few-Shot Learning; Class-Incremental Learning;
D O I
10.1145/3534678.3539280
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
With the tremendous expansion of graphs data, node classification shows its great importance in many real-world applications. Existing graph neural network based methods mainly focus on classifying unlabeled nodes within fixed classes with abundant labeling. However, in many practical scenarios, graph evolves with emergence of new nodes and edges. Novel classes appear incrementally along with few labeling due to its newly emergence or lack of exploration. In this paper, we focus on this challenging but practical graph few-shot class-incremental learning (GFSCIL) problem and propose a novel method called Geometer. Instead of replacing and retraining the fully connected neural network classifer, Geometer predicts the label of a node by finding the nearest class prototype. Prototype is a vector representing a class in the metric space. With the pop-up of novel classes, Geometer learns and adjusts the attention-based prototypes by observing the geometric proximity, uniformity and separability. Teacher-student knowledge distillation and biased sampling are further introduced to mitigate catastrophic forgetting and unbalanced labeling problem respectively. Experimental results on four public datasets demonstrate that Geometer achieves a substantial improvement of 9.46% to 27.60% over state-of-the-art methods.
引用
收藏
页码:1152 / 1161
页数:10
相关论文
共 32 条
[1]  
Bo DY, 2021, AAAI CONF ARTIF INTE, V35, P3950
[2]  
Bojchevski A., 2018, ICLR
[3]   End-to-End Incremental Learning [J].
Castro, Francisco M. ;
Marin-Jimenez, Manuel J. ;
Guil, Nicolas ;
Schmid, Cordelia ;
Alahari, Karteek .
COMPUTER VISION - ECCV 2018, PT XII, 2018, 11216 :241-257
[4]   Semantic-aware Knowledge Distillation for Few-Shot Class-Incremental Learning [J].
Cheraghian, Ali ;
Rahman, Shafin ;
Fang, Pengfei ;
Roy, Soumava Kumar ;
Petersson, Lars ;
Harandi, Mehrtash .
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, :2534-2543
[5]   Graph Prototypical Networks for Few-shot Learning on Attributed Networks [J].
Ding, Kaize ;
Wang, Jianling ;
Li, Jundong ;
Shu, Kai ;
Liu, Chenghao ;
Liu, Huan .
CIKM '20: PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, 2020, :295-304
[6]  
Finn C, 2017, PR MACH LEARN RES, V70
[7]  
Goodfellow I J, 2014, 2nd international conference on learning representations, ICLR 2014, Banff, AB. conference track proceedings
[8]  
Hinton G., 2015, arXiv
[9]   Learning a Unified Classifier Incrementally via Rebalancing [J].
Hou, Saihui ;
Pan, Xinyu ;
Loy, Chen Change ;
Wang, Zilei ;
Lin, Dahua .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :831-839
[10]  
Hou Y., 2020, INT C LEARN REPR