Graph classification method based on graph pooling contrast learning

doi:10.11772/j.issn.1001-9081.2023101526

Abstract

Abstract:

In the tasks of graph classification， the graph embedding representations obtained by the existing dropped nodes based graph pooling algorithms do not effectively utilize the implicit information in the dropped nodes and the node information between graphs. Meanwhile， traditional methods do not learn the graph embedding separately， thereby limiting some of its performance in graph classification tasks. To address the shortcomings of traditional methods， a Graph classification method based on graph Pooling Contrast Learning （GPCL） was proposed to effectively utilize the dropped node information. Firstly， the graph attention mechanism was utilized to learn the corresponding attention score for each node， and the nodes were sorted based on the attention scores and the nodes with lower scores were dropped out. Then， the nodes retained in the graph were treated as positive samples， while the dropped nodes from other graphs were treated as negative samples， the embedding representation of the graph was considered as the target node， and the pairwise similarity scores were calculated for contrast learning. Experimental results demonstrate that on D&D （Dobson PD-Doig AJ）， MUTAG， PROTEINS， and IMDB-B datasets， GPCL improves the accuracy in graph classification tasks by 5.79， 15.54， 5.42， and 1.75 percentage points respectively compared to the method using attention mechanism with hierarchical pooling alone. It is verified that GPCL enhances the utilization of inter-graph information effectively and performances well in graph classification tasks.

Key words: graph classification, graph contrast learning, graph pooling, graph neural network, unsupervised learning

摘要：

在图分类任务中，现有的利用丢弃节点的图池化算法得到的图嵌入表示没有有效地利用丢弃节点蕴含的信息和图间节点信息，同时传统方法也没有针对图嵌入进行单独学习，限制了它在图分类任务上的部分性能。为克服上述传统方法的不足，提出一种有效利用丢弃节点信息的图嵌入方法——基于图池化对比学习的图分类方法（GPCL）。首先，利用图注意力机制学习每个节点相应的注意力分数，且根据注意力分数对节点进行排序并丢弃分数较低的节点；其次，将本图保留的节点作为正样本，将其他图被丢弃的部分节点作为负样本，而将图的嵌入表达作为目标节点，两两计算相似性分数，从而进行对比学习。实验结果表明：在D&D （Dobson PD-Doig AJ）、MUTAG、PROTEINS和IMDB-B数据集上，相较于仅使用注意力机制和分层池化的方法，GPCL在图分类任务上的准确率分别提升了5.79、15.54、5.42和1.75个百分点，验证了GPCL充分提高了图间信息的利用率，在图分类任务上表现良好。

关键词: 图分类, 图对比学习, 图池化, 图神经网络, 无监督学习

CLC Number:

TP183

Nengbing HU, Biao CAI, Xu LI, Danhua CAO. Graph classification method based on graph pooling contrast learning[J]. Journal of Computer Applications, 2024, 44(11): 3327-3334.

胡能兵, 蔡彪, 李旭, 曹旦华. 基于图池化对比学习的图分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(11): 3327-3334.

Figures/Tables 10

References 37

1	LAZER D， PENTLAND A， ADAMIC L， et al. Computational social science［J］. Science， 2009， 323（5915）： 721-723.
2	DUVENAUD D K， MACLAURIN D， AGUILERA-IPARRAGUIRRE J， et al. Convolutional networks on graphs for learning molecular fingerprints［C］// Proceedings of the 28th International Conference on Neural Information Processing Systems — Volume 2. Cambridge： MIT Press， 2015： 2224-2232.
3	BEPLER T， BERGER B. Learning the protein language： evolution， structure， and function［J］. Cell Systems， 2021， 12（6）： 654-669.
4	SHERVASHIDZE N， SCHWEITZER P， JAN E， et al. Weisfeiler-Lehman graph kernels［J］. The Journal of Machine Learning Research， 2011， 12（2011）：2539-2561.
5	BORGWARDT K M， KRIEGEL H P. Shortest-path kernels on graphs［C］// Proceedings of the 5th IEEE International Conference on Data Mining. Piscataway： IEEE， 2005： 1-8.
6	SHERVASHIDZE N， VISHWANATHAN S V N， PETRI T H， et al. Efficient graphlet kernels for large graph comparison［C］// Proceedings of the 12th International Conference on Artificial Intelligence and Statistics. New York： JMLR， 2009： 488-495.
7	MA Y， WANG S， AGGARWAL C C， et al. Graph convolutional networks with EigenPooling［C］// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2019： 723-731.
8	MAGNER A， BARANWAL M， HERO A O. Fundamental limits of deep graph convolutional networks for graph classification［J］. IEEE Transactions on Information Theory， 2022， 68（5）： 3218-3233.
9	BAEK J， KANG M， HWANG S J. Accurate learning of graph representations with graph multiset pooling［EB/OL］. ［2023-09-15］. .
10	LEE J， LEE I， KANG J. Self-attention graph pooling［C］// Proceedings of the 36th International Conference on Machine Learning. New York： JMLR， 2019： 3734-3743.
11	RANJAN E， SANYAL S， TALUKDAR P. ASAP： adaptive structure aware pooling for learning hierarchical graph representations［C］// Proceedings of the 34th AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2020： 5470-5477.
12	YUAN H， JI S. StructPool： structured graph pooling via conditional random fields［EB/OL］. ［2023-05-25］..
13	NEUHAUS M， BUNKE H. Automatic learning of cost functions for graph edit distance［J］. Information Sciences， 2007， 177（1）： 239-247.
14	MORRIS C， RITZERT M， FEY M， et al. Weisfeiler and Leman go neural： higher-order graph neural networks［C］// Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2019： 4602-4609.
15	王兆慧，沈华伟，曹婍，等.图分类研究综述［J］.软件学报，2022，33（1）：171-192.
	WANG Z H， SHEN H W， CAO Q， et al. Survey on graph classification［J］. Journal of Software， 2022， 33（1）： 171-192.
16	PASA L， NAVARIN N， ERB W， et al. Empowering simple graph convolutional networks［J］. IEEE Transactions on Neural Networks and Learning Systems， 2024， 35（4）： 4385-4399.
17	CAI B， ZHU X， QIN Y. Parameters optimization of hybrid strategy recommendation based on particle swarm algorithm［J］. Expert Systems with Applications， 2021， 168： No.114388.
18	CAI B， YANG X， HUANG Y， et al. A triangular personalized recommendation algorithm for improving diversity［J］. Discrete Dynamics in Nature and Society， 2018， 2018（1）： No.3162068.
19	WU L， XIA J， GAO Z， et al. GraphMixup： improving class-imbalanced node classification by reinforcement mixup and self-supervised context prediction［C］// Proceedings of the 2022 Joint European Conference on Machine Learning and Knowledge Discovery in Databases， LNCS 13716. Cham： Springer， 2023： 519-535.
20	KIPF T N， WELLING M. Semi-supervised classification with graph convolutional networks［EB/OL］. ［2023-10-13］..
21	VINYALS O， BENGIO S， KUDLUR M. Order matters： sequence to sequence for sets［EB/OL］. ［2023-09-01］..
22	ZHANG M， CUI Z， NEUMANN M， et al. An end-to-end deep learning architecture for graph classification［C］// Proceedings of the 32nd AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2018： 4438-4445.
23	YING R， YOU J， MORRIS C， et al. Hierarchical graph representation learning with differentiable pooling［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2018： 4805-4815.
24	VELIČKOVIĆ P， CUCURULL G， CASANOVA A， et al. Graph attention networks［EB/OL］. ［2023-08-04］..
25	YAN Z， CAO W， JI J. Social behavior prediction with graph U‑Net+［J］. Discover Internet of Things， 2021， 1： No.18.
26	LeCUN Y. A theoretical framework for back-propagation［C］// Proceedings of the 1988 Connectionist Models Summer School. San Mateo： Morgan Kaufmann Publishers， Inc.， 1989： 21-28.
27	CANGEA C， VELIČKOVIĆ P， JOVANOVIĆ N， et al. Towards sparse hierarchical graph classifiers［EB/OL］. ［2023-07-25］..
28	FEY M， LENSSEN J E. Fast graph representation learning with PyTorch Geometric［EB/OL］. ［2023-07-06］..
29	PASZKE A， GROSS S， CHINTALA S， et al. Automatic differentiation in PyTorch［EB/OL］. ［2023-06-19］..
30	DOBSON P D， DOIG A J. Distinguishing enzyme structures from non-enzymes without alignments［J］. Journal of Molecular Biology， 2003， 330（4）： 771-783.
31	BORGWARDT K M， ONG C S， SCHÖNAUER S， et al. Protein function prediction via graph kernels［J］. Bioinformatics， 2005， 21（S1）： i47-i56.
32	DEBNATH A K， DE COMPADRE R L L， DEBNATH G， et al. Structure-activity relationship of mutagenic aromatic and heteroaromatic nitro compounds. Correlation with molecular orbital energies and hydrophobicity［J］. Journal of Medicinal Chemistry， 1991， 34（2）： 786-797.
33	YANARDAG P， VISHWANATHAN S V N. Deep graph kernels［C］// Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2015：1365-1374.
34	SHCHUR O， MUMME M， BOJCHEVSKI A， et al. Pitfalls of graph neural network evaluation［EB/OL］. ［2024-01-30］..
35	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. ［2023-08-17］..
36	BIANCHI F M， GRATTAROLA D， ALIPPI C. Spectral clustering with graph neural networks for graph pooling［C］// Proceedings of the 37th International Conference on Machine Learning. New York： JMLR， 2020： 874-883.
37	WU J， CHEN X， XU K， et al. Structural entropy guided graph hierarchical pooling［C］// Proceedings of the 39th International Conference on Machine Learning. New York： JMLR， 2022： 24017-24030.
	This work is partially supported byFoundation: National Natural Science Foundation of China （61802034）.

数据集	图数	类别数	平均节点数	平均边数
D&D	1 178	2	284.32	1 431.30
MUTAG	188	2	17.93	19.79
PROTEINS	1 113	2	39.06	72.82

数据集	图数	类别数	平均节点数	平均边数
D&D	1 178	2	284.32	1 431.30
MUTAG	188	2	17.93	19.79
PROTEINS	1 113	2	39.06	72.82

数据集	图数	类别数	平均节点数	平均边数
IMDB-B	1 000	2	19.77	96.53
IMDB-M	1 500	3	13.00	65.94

数据集	图数	类别数	平均节点数	平均边数
IMDB-B	1 000	2	19.77	96.53
IMDB-M	1 500	3	13.00	65.94

超参数	取值范围
学习率	0.000 1，0.000 5，0.001
隐藏层维度	64，128
池化率	0.4，0.5，0.6