Graph Retention Networks for Dynamic Graphs
Not on the map. Why?
Abstract (arXiv)
In this paper, we propose Graph Retention Networks (GRNs) as a unified architecture for deep learning on dynamic graphs. The GRN extends the concept of retention into dynamic graph data as graph retention, equipping the model with three key computational paradigms: parallelizable training, low-cost O(1) inference, and long-term chunkwise training. This architecture achieves an optimal balance between efficiency, effectiveness, and scalability. Extensive experiments on benchmark datasets demonstrate its strong performance in both edge-level prediction and node-level classification tasks with significantly reduced training latency, lower GPU memory overhead, and improved inference throughput by up to 86.7x compared to SOTA baselines. The proposed GRN architecture achieves competitive performance across diverse dynamic graph benchmarks, demonstrating its adaptability to a wide range of tasks.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 26 of its 48 references that are in the map. First posted on arXiv 18 Nov 2024.
Builds on
26 of 48 references.
References · 26
- 2017Attention Is All You Needcited by 59,932
- 2018Group Normalizationcited by 1,514
- 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,125
- 2019Graph WaveNet for Deep Spatial-Temporal Graph Modelingcited by 820
- 2020Temporal Graph Networks for Deep Learning on Dynamic Graphscited by 521
- 2016Interaction Networks for Learning about Objects, Relations and Physicscited by 503
- 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
- 2014Memory Networkscited by 354
- 2021Inductive Representation Learning in Temporal Networks via Causal Anonymous Walkscited by 188
- 2023Towards Better Dynamic Graph Learning: New Architecture and Unified Librarycited by 154
14 more
- 2017A Survey on Network Embeddingcited by 142
- 2023Do We Really Need Complicated Model Architectures For Temporal Networks?cited by 141
- 2022Towards Better Evaluation for Dynamic Link Predictioncited by 120
- 2017Dynamic Graph Convolutional Networkscited by 83
- 2021A Survey on Embedding Dynamic Graphscited by 60
- 2024A Comprehensive Survey of Dynamic Graph Neural Networks: Models, Frameworks, Benchmarks, Experiments and Challengescited by 29
- 2020Evolutionary dynamics of higher-order interactions in social networkscited by 25
- 2022Scaling Up Dynamic Graph Representation Learning via Spiking Neural Networkscited by 24
- 2024On the Feasibility of Simple Transformer for Dynamic Graph Modelingcited by 22
- 2023The Evolution of Distributed Systems for Graph Neural Networks and their Origin in Graph Processing and Deep Learning: A Surveycited by 13
- 2019Practice of Streaming Processing of Dynamic Graphs: Concepts, Models, and Systemscited by 12
- 2023DistTGL: Distributed Memory-Based Temporal Graph Neural Network Trainingcited by 12
- 2024Towards Adaptive Neighborhood for Advancing Temporal Interaction Graph Modelingcited by 8
- 2024Exploiting Individual Graph Structures to Enhance Ecological Momentary Assessment (EMA) Forecastingcited by 1
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.