TASER: Temporal Adaptive Sampling for Fast and Accurate Dynamic Graph Representation Learning
Not on the map. Why?
Abstract (arXiv)
Recently, Temporal Graph Neural Networks (TGNNs) have demonstrated state-of-the-art performance in various high-impact applications, including fraud detection and content recommendation. Despite the success of TGNNs, they are prone to the prevalent noise found in real-world dynamic graphs like time-deprecated links and skewed interaction distribution. The noise causes two critical issues that significantly compromise the accuracy of TGNNs: (1) models are supervised by inferior interactions, and (2) noisy input induces high variance in the aggregated messages. However, current TGNN denoising techniques do not consider the diverse and dynamic noise pattern of each node. In addition, they also suffer from the excessive mini-batch generation overheads caused by traversing more neighbors. We believe the remedy for fast and accurate TGNNs lies in temporal adaptive sampling. In this work, we propose TASER, the first adaptive sampling method for TGNNs optimized for accuracy, efficiency, and scalability. TASER adapts its mini-batch selection based on training dynamics and temporal neighbor selection based on the contextual, structural, and temporal properties of past interactions. To alleviate the bottleneck in mini-batch generation, TASER implements a pure GPU-based temporal neighbor finder and a dedicated GPU feature cache. We evaluate the performance of TASER using two state-of-the-art backbone TGNNs. On five popular datasets, TASER outperforms the corresponding baselines by an average of 2.3% in Mean Reciprocal Rank (MRR) while achieving an average of 5.1x speedup in training time.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 34 of its 43 references that are in the map. First posted on arXiv 8 Feb 2024.
Builds on
34 of 43 references.
References · 34
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2017Graph Attention Networkscited by 8,854
- 2017Inductive Representation Learning on Large Graphscited by 6,284
- 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
- 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
- 2021How Attentive are Graph Attention Networks?cited by 818
- 2019GraphSAINT: Graph Sampling Based Inductive Learning Methodcited by 618
- 2018FastGCN: Fast Learning with Graph Convolutional Networks via Importance Samplingcited by 583
- 2020Temporal Graph Networks for Deep Learning on Dynamic Graphscited by 521
- 2020Inductive Representation Learning on Temporal Graphscited by 454
- 2019EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphscited by 443
- 2019Predicting Dynamic Embedding Trajectory in Temporal Interaction Networkscited by 443
22 more
- 2018Adaptive Sampling Towards Fast Graph Representation Learningcited by 180
- 2019Layer-Dependent Importance Sampling for Training Deep and Large Graph Convolutional Networkscited by 177
- 2023Do We Really Need Complicated Model Architectures For Temporal Networks?cited by 141
- 2022Towards Better Evaluation for Dynamic Link Predictioncited by 120
- 2022TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphscited by 59
- 2021Dynamic Graph Neural Networks for Sequential Recommendationcited by 57
- 2020Minimal Variance Sampling with Provable Guarantees for Fast Training of Graph Neural Networkscited by 54
- 2022Neighborhood-aware Scalable Temporal Network Representation Learningcited by 54
- 2020Bandit Samplers for Training Graph Neural Networkscited by 30
- 2020Accelerating Graph Sampling for Graph Machine Learning using GPUscited by 29
- 2021Global Neighbor Sampling for Mixed CPU-GPU Training on Giant Graphscited by 22
- 2017Safe Adaptive Importance Samplingcited by 19
- 2021A Biased Graph Neural Network Sampler with Near-Optimal Regretcited by 17
- 2020C-SAW: A Framework for Graph Sampling and Random Walk on GPUscited by 13
- 2023DistTGL: Distributed Memory-Based Temporal Graph Neural Network Trainingcited by 12
- 2023Quiver: Supporting GPUs for Low-Latency, High-Throughput GNN Serving with Workload Awarenesscited by 11
- 2021Time-Aware Neighbor Sampling for Temporal Graph Networkscited by 9
- 2018Gradient-based Sampling: An Adaptive Importance Sampling for Least-squarescited by 8
- 2021Graph Neural Network Training with Data Tieringcited by 8
- 2023Less Can Be More: Unsupervised Graph Pruning for Large-scale Dynamic Graphscited by 7
- 2023Temporal Graph Representation Learning with Adaptive Augmentation Contrastivecited by 5
- 2021SeDyT: A General Framework for Multi-Step Event Forecasting via Sequence Modeling on Dynamic Entity Embeddingscited by 3
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.