TransformerG2G: Adaptive time-stepping for learning temporal graph embeddings using transformers
Document Type
Article
Publication Date
4-1-2024
Abstract
Dynamic graph embedding has emerged as a very effective technique for addressing diverse temporal graph analytic tasks (i.e., link prediction, node classification, recommender systems, anomaly detection, and graph generation) in various applications. Such temporal graphs exhibit heterogeneous transient dynamics, varying time intervals, and highly evolving node features throughout their evolution. Hence, incorporating long-range dependencies from the historical graph context plays a crucial role in accurately learning their temporal dynamics. In this paper, we develop a graph embedding model with uncertainty quantification, TransformerG2G, by exploiting the advanced transformer encoder to first learn intermediate node representations from its current state (t) and previous context (over timestamps [t−1,t−l], l is the length of context). Moreover, we employ two projection layers to generate lower-dimensional multivariate Gaussian distributions as each node's latent embedding at timestamp t. We consider diverse benchmarks with varying levels of “novelty” as measured by the TEA (Temporal Edge Appearance) plots. Our experiments demonstrate that the proposed TransformerG2G model outperforms conventional multi-step methods and our prior work (DynG2G) in terms of both link prediction accuracy and computational efficiency, especially for high degree of novelty. Furthermore, the learned time-dependent attention weights across multiple graph snapshots reveal the development of an automatic adaptive time stepping enabled by the transformer. Importantly, by examining the attention weights, we can uncover temporal dependencies, identify influential elements, and gain insights into the complex interactions within the graph structure. For example, we identified a strong correlation between attention weights and node degree at the various stages of the graph topology evolution.
Identifier
85181090464 (Scopus)
Publication Title
Neural Networks
External Full Text Location
https://doi.org/10.1016/j.neunet.2023.12.040
e-ISSN
18792782
ISSN
08936080
PubMed ID
38159511
Volume
172
Grant
N00014-22-1-2795
Fund Ref
U.S. Department of Energy
Recommended Citation
Varghese, Alan John; Bora, Aniruddha; Xu, Mengjia; and Karniadakis, George Em, "TransformerG2G: Adaptive time-stepping for learning temporal graph embeddings using transformers" (2024). Faculty Publications. 536.
https://digitalcommons.njit.edu/fac_pubs/536
