Unsupervised Graph Representation Learning Beyond Aggregated View

Unsupervised graph representation learning aims to condense graph information into dense vector embeddings to support various downstream tasks. To achieve this goal, existing UGRL approaches mainly adopt the message-passing mechanism to simultaneously incorporate graph topology and node attribute wi...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on knowledge and data engineering Ročník 36; číslo 12; s. 9504 - 9516
Hlavní autoři: Zhou, Jian, Li, Jiasheng, Kuang, Li, Gui, Ning
Médium: Journal Article
Jazyk:angličtina
Vydáno: IEEE 01.12.2024
Témata:
Benchmark testing
Data mining
Decoding
dual-view autoencoder
Encoding
Graph representation learning
Network topology
self-supervised
Synthetic data
Topology
topology and attribute features
ISSN:1041-4347, 1558-2191
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Unsupervised graph representation learning aims to condense graph information into dense vector embeddings to support various downstream tasks. To achieve this goal, existing UGRL approaches mainly adopt the message-passing mechanism to simultaneously incorporate graph topology and node attribute with an aggregated view. However, recent research points out that this direct aggregation may lead to issues such as over-smoothing and/or topology distortion, as topology and node attribute of totally different semantics. To address this issue, this paper proposes a novel Graph Dual-view AutoEncoder framework (GDAE) which introduces the node-wise view for an individual node beyond the traditional aggregated view for aggregation of connected nodes. Specifically, the node-wise view captures the unique characteristics of individual node through a decoupling design, i.e., topology encoding by multi-steps random walk while preserving node-wise individual attribute. Meanwhile, the aggregated view aims to better capture the collective commonality among long-range nodes through an enhanced strategy, i.e., topology masking then attribute aggregation. Extensive experiments on 5 synthetic and 11 real-world benchmark datasets demonstrate that GDAE achieves the best results with up to 49.5% and 21.4% relative improvement in node degree prediction and cut-vertex detection tasks and remains top in node classification and link prediction tasks.
ISSN:1041-4347
1558-2191
DOI:10.1109/TKDE.2024.3418576