SimGRL: a simple self-supervised graph representation learning framework via triplets
Abstract Recently, graph contrastive learning (GCL) has achieved remarkable performance in graph representation learning. However, existing GCL methods usually follow a dual-channel encoder network (i.e., Siamese networks), which adds to the complexity of the network architecture. Additionally, thes...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-02-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-023-00997-6 |
| _version_ | 1797674976353976320 |
|---|---|
| author | Da Huang Fangyuan Lei Xi Zeng |
| author_facet | Da Huang Fangyuan Lei Xi Zeng |
| author_sort | Da Huang |
| collection | DOAJ |
| description | Abstract Recently, graph contrastive learning (GCL) has achieved remarkable performance in graph representation learning. However, existing GCL methods usually follow a dual-channel encoder network (i.e., Siamese networks), which adds to the complexity of the network architecture. Additionally, these methods overly depend on varied data augmentation techniques, corrupting graph information. Furthermore, they are heavily reliant on large quantities of negative nodes for each object node, which requires tremendous memory costs. To address these issues, we propose a novel and simple graph representation learning framework, named SimGRL. Firstly, our proposed network architecture only contains one encoder based on a graph neural network instead of a dual-channel encoder, which simplifies the network architecture. Then we introduce a distributor to generate triplets to obtain the contrastive views between nodes and their neighbors, avoiding the need for data augmentations. Finally, we design a triplet loss based on adjacency information in graphs that utilizes only one negative node for each object node, reducing memory overhead significantly. Extensive experiments demonstrate that SimGRL achieves competitive performance on node classification and graph classification tasks, especially in terms of running time and memory overhead. |
| first_indexed | 2024-03-11T22:08:00Z |
| format | Article |
| id | doaj.art-7a684ff7b1034a2c956fc4c5ed0a7175 |
| institution | Directory Open Access Journal |
| issn | 2199-4536 2198-6053 |
| language | English |
| last_indexed | 2024-03-11T22:08:00Z |
| publishDate | 2023-02-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj.art-7a684ff7b1034a2c956fc4c5ed0a71752023-09-24T11:35:38ZengSpringerComplex & Intelligent Systems2199-45362198-60532023-02-01955049506210.1007/s40747-023-00997-6SimGRL: a simple self-supervised graph representation learning framework via tripletsDa Huang0Fangyuan Lei1Xi Zeng2School of Electronic and Information, Guangdong Polytechnic Normal UniversitySchool of Electronic and Information, Guangdong Polytechnic Normal UniversitySchool of Electronic and Information, Guangdong Polytechnic Normal UniversityAbstract Recently, graph contrastive learning (GCL) has achieved remarkable performance in graph representation learning. However, existing GCL methods usually follow a dual-channel encoder network (i.e., Siamese networks), which adds to the complexity of the network architecture. Additionally, these methods overly depend on varied data augmentation techniques, corrupting graph information. Furthermore, they are heavily reliant on large quantities of negative nodes for each object node, which requires tremendous memory costs. To address these issues, we propose a novel and simple graph representation learning framework, named SimGRL. Firstly, our proposed network architecture only contains one encoder based on a graph neural network instead of a dual-channel encoder, which simplifies the network architecture. Then we introduce a distributor to generate triplets to obtain the contrastive views between nodes and their neighbors, avoiding the need for data augmentations. Finally, we design a triplet loss based on adjacency information in graphs that utilizes only one negative node for each object node, reducing memory overhead significantly. Extensive experiments demonstrate that SimGRL achieves competitive performance on node classification and graph classification tasks, especially in terms of running time and memory overhead.https://doi.org/10.1007/s40747-023-00997-6Graph representation learningGraph neural networksSelf-supervised learningTriplet lossNode classificationGraph classification |
| spellingShingle | Da Huang Fangyuan Lei Xi Zeng SimGRL: a simple self-supervised graph representation learning framework via triplets Complex & Intelligent Systems Graph representation learning Graph neural networks Self-supervised learning Triplet loss Node classification Graph classification |
| title | SimGRL: a simple self-supervised graph representation learning framework via triplets |
| title_full | SimGRL: a simple self-supervised graph representation learning framework via triplets |
| title_fullStr | SimGRL: a simple self-supervised graph representation learning framework via triplets |
| title_full_unstemmed | SimGRL: a simple self-supervised graph representation learning framework via triplets |
| title_short | SimGRL: a simple self-supervised graph representation learning framework via triplets |
| title_sort | simgrl a simple self supervised graph representation learning framework via triplets |
| topic | Graph representation learning Graph neural networks Self-supervised learning Triplet loss Node classification Graph classification |
| url | https://doi.org/10.1007/s40747-023-00997-6 |
| work_keys_str_mv | AT dahuang simgrlasimpleselfsupervisedgraphrepresentationlearningframeworkviatriplets AT fangyuanlei simgrlasimpleselfsupervisedgraphrepresentationlearningframeworkviatriplets AT xizeng simgrlasimpleselfsupervisedgraphrepresentationlearningframeworkviatriplets |