# Localised Adaptive Spatial-Temporal Graph Neural Network

arXiv: https://arxiv.org/abs/2306.06930 · HTML: https://mlsanity.com/paper/2306.06930/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 31 references that are in the map.
- First posted on arXiv 12 Jun 2023.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting](https://mlsanity.com/paper/1709.04875/) (2017)
- [Graph WaveNet for Deep Spatial-Temporal Graph Modeling](https://mlsanity.com/paper/1906.00121/) (2019)
- [Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting](https://mlsanity.com/paper/1907.00235/) (2019)
- [Learning Sparse Neural Networks through L0 Regularization](https://mlsanity.com/paper/1712.01312/) (2017)
- [Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting](https://mlsanity.com/paper/2007.02842/) (2020)
- [Structured Sequence Modeling with Graph Convolutional Recurrent Networks](https://mlsanity.com/paper/1612.07659/) (2016)
- [Graph Neural Controlled Differential Equations for Traffic Forecasting](https://mlsanity.com/paper/2112.03558/) (2021)
- [A Unified Lottery Ticket Hypothesis for Graph Neural Networks](https://mlsanity.com/paper/2102.06790/) (2021)
- [Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting](https://mlsanity.com/paper/2105.04100/) (2021)
- [STG2Seq: Spatial-temporal Graph to Sequence Model for Multi-step Passenger Demand Forecasting](https://mlsanity.com/paper/1905.10069/) (2019)
- [Spatio-Temporal Data Mining: A Survey of Problems and Methods](https://mlsanity.com/paper/1711.04710/) (2017)
- [Early-Bird GCNs: Graph-Network Co-Optimization Towards More Efficient GCN Training and Inference via Drawing Early-Bird Lottery Tickets](https://mlsanity.com/paper/2103.00794/) (2021)
- [Dissecting Ethereum Blockchain Analytics: What We Learn from Topology and Geometry of Ethereum Graph](https://mlsanity.com/paper/1912.10105/) (2019)
- [Combating Distribution Shift for Accurate Time Series Forecasting via Hypernetworks](https://mlsanity.com/paper/2202.10808/) (2022)

## Built on by (most cited first, 14 of 14)

- [Multi-scale Traffic Pattern Bank for Cross-city Few-shot Traffic Forecasting](https://mlsanity.com/paper/2402.00397/) (2024)
- [Frequency Enhanced Pre-training for Cross-city Few-shot Traffic Forecasting](https://mlsanity.com/paper/2406.02614/) (2024)
- [Scalable Transformer for High Dimensional Multivariate Time Series Forecasting](https://mlsanity.com/paper/2408.04245/) (2024)
- [RAM: Replace Attention with MLP for Efficient Multivariate Time Series Forecasting](https://mlsanity.com/paper/2410.24023/) (2024)
- [Pre-Training Identification of Graph Winning Tickets in Adaptive Spatial-Temporal Graph Neural Networks](https://mlsanity.com/paper/2406.08287/) (2024)
- [Dynamic Localisation of Spatial-Temporal Graph Neural Network](https://mlsanity.com/paper/2501.04239/) (2025)
- [GraphSparseNet: a Novel Method for Large Scale Traffic Flow Prediction](https://mlsanity.com/paper/2502.19823/) (2025)
- [Efficient Traffic Forecasting on Large-Scale Road Network by Regularized Adaptive Graph Convolution](https://mlsanity.com/paper/2506.07179/) (2025)
- [A General ReLearner: Empowering Spatiotemporal Prediction by Re-learning Input-label Residual](https://mlsanity.com/paper/2602.02563/) (2026)
- [Minimalist Traffic Prediction: Linear Layer Is All You Need](https://mlsanity.com/paper/2308.10276/) (2023)
- [Towards Pattern-aware Data Augmentation for Temporal Knowledge Graph Completion](https://mlsanity.com/paper/2501.00252/) (2024)
- [SPAT: Sensitivity-based Multihead-attention Pruning on Time Series Forecasting Models](https://mlsanity.com/paper/2505.08768/) (2025)
- [ADMFormer: An Adaptive-Decomposition Transformer with Time-Varying Masked Spatial Attention for Traffic Forecasting](https://mlsanity.com/paper/2605.25543/) (2026)
- [PHGNet: Prototype-Guided Hypergraph Construction for Heterogeneous Spatiotemporal Forecasting](https://mlsanity.com/paper/2605.25554/) (2026)

## Abstract (arXiv)

Spatial-temporal graph models are prevailing for abstracting and modelling spatial and temporal dependencies. In this work, we ask the following question: whether and to what extent can we localise spatial-temporal graph models? We limit our scope to adaptive spatial-temporal graph neural networks (ASTGNNs), the state-of-the-art model architecture. Our approach to localisation involves sparsifying the spatial graph adjacency matrices. To this end, we propose Adaptive Graph Sparsification (AGS), a graph sparsification algorithm which successfully enables the localisation of ASTGNNs to an extreme extent (fully localisation). We apply AGS to two distinct ASTGNN architectures and nine spatial-temporal datasets. Intriguingly, we observe that spatial graphs in ASTGNNs can be sparsified by over 99.5% without any decline in test accuracy. Furthermore, even when ASTGNNs are fully localised, becoming graph-less and purely temporal, we record no drop in accuracy for the majority of tested datasets, with only minor accuracy deterioration observed in the remaining datasets. However, when the partially or fully localised ASTGNNs are reinitialised and retrained on the same data, there is a considerable and consistent drop in accuracy. Based on these observations, we reckon that textit(i) in the tested data, the information provided by the spatial dependencies is primarily included in the information provided by the temporal dependencies and, thus, can be essentially ignored for inference; and textit(ii) although the spatial dependencies provide redundant information, it is vital for the effective training of ASTGNNs and thus cannot be ignored during training. Furthermore, the localisation of ASTGNNs holds the potential to reduce the heavy computation overhead required on large-scale spatial-temporal data and further enable the distributed deployment of ASTGNNs.