TIGER: Temporal Interaction Graph Embedding with Restarts

TL;DR

TIGER is a novel temporal interaction graph embedding model that enables parallel processing and improves representation freshness by restarting at multiple timestamps with a dual memory system.

Contribution

The paper introduces TIGER, a model allowing restarts at any timestamp and parallelization, with a dual memory module to enhance neighborhood information utilization.

Findings

TIGER achieves superior accuracy on multiple datasets.

TIGER significantly reduces training time through parallelization.

The dual memory module alleviates staleness in node representations.

Abstract

Temporal interaction graphs (TIGs), consisting of sequences of timestamped interaction events, are prevalent in fields like e-commerce and social networks. To better learn dynamic node embeddings that vary over time, researchers have proposed a series of temporal graph neural networks for TIGs. However, due to the entangled temporal and structural dependencies, existing methods have to process the sequence of events chronologically and consecutively to ensure node representations are up-to-date. This prevents existing models from parallelization and reduces their flexibility in industrial applications. To tackle the above challenge, in this paper, we propose TIGER, a TIG embedding model that can restart at any timestamp. We introduce a restarter module that generates surrogate representations acting as the warm initialization of node representations. By restarting from multiple timestamps simultaneously, we divide the sequence into multiple chunks and naturally enable the parallelization of the model. Moreover, in contrast to previous models that utilize a single memory unit, we introduce a dual memory module to better exploit neighborhood information and alleviate the staleness problem. Extensive experiments on four public datasets and one industrial dataset are conducted, and the results verify both the effectiveness and the efficiency of our work.