Cabanac et al. (2022)
paperAuthors
Credibility Rating
Good quality. Reputable source with community review or editorial standards, but less rigorous than peer-reviewed venues.
Rating inherited from publication venue: arXiv
A machine learning paper on spatio-temporal traffic prediction using graph neural networks; relevant to AI safety as an example of complex ML system design in safety-critical transportation domains.
Paper Details
Metadata
Abstract
Accurate traffic prediction is a challenging task in intelligent transportation systems because of the complex spatio-temporal dependencies in transportation networks. Many existing works utilize sophisticated temporal modeling approaches to incorporate with graph convolution networks (GCNs) for capturing short-term and long-term spatio-temporal dependencies. However, these separated modules with complicated designs could restrict effectiveness and efficiency of spatio-temporal representation learning. Furthermore, most previous works adopt the fixed graph construction methods to characterize the global spatio-temporal relations, which limits the learning capability of the model for different time periods and even different data scenarios. To overcome these limitations, we propose an automated dilated spatio-temporal synchronous graph network, named Auto-DSTSGN for traffic prediction. Specifically, we design an automated dilated spatio-temporal synchronous graph (Auto-DSTSG) module to capture the short-term and long-term spatio-temporal correlations by stacking deeper layers with dilation factors in an increasing order. Further, we propose a graph structure search approach to automatically construct the spatio-temporal synchronous graph that can adapt to different data scenarios. Extensive experiments on four real-world datasets demonstrate that our model can achieve about 10% improvements compared with the state-of-art methods. Source codes are available at https://github.com/jinguangyin/Auto-DSTSGN.
Summary
This paper proposes Auto-DSTSGN, an automated dilated spatio-temporal synchronous graph network for traffic prediction in intelligent transportation systems. The key innovation is an automated graph structure search approach that dynamically constructs the spatio-temporal graph to adapt to different data scenarios, rather than using fixed graph construction methods. The model uses dilated layers with increasing dilation factors to capture both short-term and long-term spatio-temporal dependencies more effectively. Experiments on four real-world datasets demonstrate approximately 10% performance improvements over state-of-the-art methods.
Cited by 1 page
| Page | Type | Quality |
|---|---|---|
| Scientific Knowledge Corruption | Risk | 91.0 |
Cached Content Preview
# Automated Dilated Spatio-Temporal Synchronous Graph Modeling for Traffic Prediction
Guangyin Jin\*,
Fuxian Li\*,
Jinlei Zhang,
Mudan Wang and
Jincai Huang
F. Li, M. Wang are with Beijing National Research Center for Information Science and Technology (BNRist), Department of Electronic Engineering, Tsinghua University, Beijing 100084, China.
E-mail: lifx19@mails.tsinghua.edu.cn, wmd20@mails.tsinghua.edu.cn
G. Jin and J. Huang are with College of Systems Engineering, National University of Defense Technology, Changsha, China.
E-mail: jinguangyin18@nudt.edu.cn, huangjincai@nudt.edu.cn
J. Zhang is with State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, China.
E-mail: zhangjinlei@bjtu.edu.cn
∗Both authors contributed equally to this research.
###### Abstract
Accurate traffic prediction is a challenging task in intelligent
transportation systems because of the complex spatio-temporal dependencies in transportation networks. Many existing works utilize sophisticated temporal modeling approaches to incorporate with graph convolution networks (GCNs) for capturing short-term and long-term spatio-temporal dependencies. However, these separated modules with complicated designs could restrict effectiveness and efficiency of spatio-temporal representation learning. Furthermore, most previous works adopt the fixed graph construction methods to characterize the global spatio-temporal relations, which limits the learning capability of the model for different time periods and even different data scenarios. To overcome these limitations, we propose an automated dilated spatio-temporal synchronous graph network, named Auto-DSTSGN for traffic prediction. Specifically, we design an automated dilated spatio-temporal synchronous graph (Auto-DSTSG) module to capture the short-term and long-term spatio-temporal correlations by stacking deeper layers with dilation factors in an increasing order. Further, we propose a graph structure search approach to automatically construct the spatio-temporal synchronous graph that can adapt to different data scenarios. Extensive experiments on four real-world datasets demonstrate that our model can achieve about 10%percent1010\\% improvements compared with the state-of-art methods. Source codes are available at https://github.com/jinguangyin/Auto-DSTSGN.
###### Index Terms:
Traffic prediction, spatio-temporal modeling, graph neural networks, automated machine learning
## I Introduction
Traffic prediction plays a basic but crucial role in intelligent transportation systems, which has been widely deployed in some online services such as navigation and ride-hailing.
Effective spatio-temporal modeling is key to obtain more precise predictions. In recent years, most works utilize spatial modules including convolutional neural networks (CNNs) and graph neural networks (GNNs), and temporal modules such as recurrent neural networks (RNNs), temporal convolution networks (TCNs) for
... (truncated, 88 KB total)dfae5307f40ea28e | Stable ID: Y2M2NTc3ND