Dynamic Spatiotemporal Graph Convolutional Neural Networks for Traffic Data Imputation with Complex Missing Patterns

• URL: http://arxiv.org/abs/2109.08357v1
• Date: Fri, 17 Sep 2021 05:47:17 GMT
• Title: Dynamic Spatiotemporal Graph Convolutional Neural Networks for Traffic Data Imputation with Complex Missing Patterns
• Authors: Yuebing Liang, Zhan Zhao, Lijun Sun
• Abstract summary: We propose a novel deep learning framework called Dynamic Spatio Graph Contemporal Networks (DSTG) to impute missing traffic data. We introduce a graph structure estimation technique to model the dynamic spatial dependencies real-time traffic information and road network structure. Our proposed model outperforms existing deep learning models in all kinds of missing scenarios and the graph structure estimation technique contributes to the model performance.
• Score: 3.9318191265352196
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Missing data is an inevitable and ubiquitous problem for traffic data collection in intelligent transportation systems. Despite extensive research regarding traffic data imputation, there still exist two limitations to be addressed: first, existing approaches fail to capture the complex spatiotemporal dependencies in traffic data, especially the dynamic spatial dependencies evolving with time; second, prior studies mainly focus on randomly missing patterns while other more complex missing scenarios are less discussed. To fill these research gaps, we propose a novel deep learning framework called Dynamic Spatiotemporal Graph Convolutional Neural Networks (DSTGCN) to impute missing traffic data. The model combines the recurrent architecture with graph-based convolutions to model the spatiotemporal dependencies. Moreover, we introduce a graph structure estimation technique to model the dynamic spatial dependencies from real-time traffic information and road network structure. Extensive experiments based on two public traffic speed datasets are conducted to compare our proposed model with state-of-the-art deep learning approaches in four types of missing patterns. The results show that our proposed model outperforms existing deep learning models in all kinds of missing scenarios and the graph structure estimation technique contributes to the model performance. We further compare our proposed model with a tensor factorization model and find distinct behaviors across different model families under different training schemes and data availability.

Related papers

• A Multi-Graph Convolutional Neural Network Model for Short-Term Prediction of Turning Movements at Signalized Intersections [0.6215404942415159]
  This study introduces a novel deep learning architecture, referred to as the multigraph convolution neural network (MGCNN) for turning movement prediction at intersections. The proposed architecture combines a multigraph structure, built to model temporal variations in traffic data, with a spectral convolution operation to support modeling the spatial variations in traffic data over the graphs. The model's ability to perform short-term predictions over 1, 2, 3, 4, and 5 minutes into the future was evaluated against four baseline state-of-the-art models.
  arXiv  Detail & Related papers  (2024-06-02T05:41:25Z)
• Spatiotemporal Implicit Neural Representation as a Generalized Traffic Data Learner [46.866240648471894]
  Spatiotemporal Traffic Data (STTD) measures the complex dynamical behaviors of the multiscale transportation system. We present a novel paradigm to address the STTD learning problem by parameterizing STTD as an implicit neural representation. We validate its effectiveness through extensive experiments in real-world scenarios, showcasing applications from corridor to network scales.
  arXiv  Detail & Related papers  (2024-05-06T06:23:06Z)
• TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts [6.831798156287652]
  We propose a novel deep learning model named TESTAM, which individually models recurring and non-recurring traffic patterns. We show that TESTAM achieves a better indication and modeling of recurring and non-recurring traffic.
  arXiv  Detail & Related papers  (2024-03-05T02:27:52Z)
• Dynamic Hypergraph Structure Learning for Traffic Flow Forecasting [35.0288931087826]
  Traffic flow forecasting aims to predict future traffic conditions on the basis of networks and traffic conditions in the past. The problem is typically solved by modeling complex-temporal correlations in traffic data using far-temporal neural networks (GNNs) Existing methods follow the paradigm of message passing that aggregates neighborhood information linearly. In this paper, we propose a model named Dynamic Hyper Structure Learning (DyHSL) for traffic flow prediction.
  arXiv  Detail & Related papers  (2023-09-21T12:44:55Z)
• STG4Traffic: A Survey and Benchmark of Spatial-Temporal Graph Neural Networks for Traffic Prediction [9.467593700532401]
  This paper provides a systematic review of graph learning strategies and commonly used graph convolution algorithms. We then conduct a comprehensive analysis of the strengths and weaknesses of recently proposed spatial-temporal graph network models. We build a study called STG4Traffic using the deep learning framework PyTorch to establish a standardized and scalable benchmark on two types of traffic datasets.
  arXiv  Detail & Related papers  (2023-07-02T06:56:52Z)
• OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning [67.07363529640784]
  We propose OpenSTL to categorize prevalent approaches into recurrent-based and recurrent-free models. We conduct standard evaluations on datasets across various domains, including synthetic moving object trajectory, human motion, driving scenes, traffic flow and forecasting weather. We find that recurrent-free models achieve a good balance between efficiency and performance than recurrent models.
  arXiv  Detail & Related papers  (2023-06-20T03:02:14Z)
• Closed-form Continuous-Depth Models [99.40335716948101]
  Continuous-depth neural models rely on advanced numerical differential equation solvers. We present a new family of models, termed Closed-form Continuous-depth (CfC) networks, that are simple to describe and at least one order of magnitude faster.
  arXiv  Detail & Related papers  (2021-06-25T22:08:51Z)
• SST-GNN: Simplified Spatio-temporal Traffic forecasting model using Graph Neural Network [2.524966118517392]
  We have designed a simplified S-temporal GNN(SST-GNN) that effectively encodes the dependency by separately aggregating different neighborhood. We have shown that our model has significantly outperformed the state-of-the-art models on three real-world traffic datasets.
  arXiv  Detail & Related papers  (2021-03-31T18:28:44Z)
• Multi-intersection Traffic Optimisation: A Benchmark Dataset and a Strong Baseline [85.9210953301628]
  Control of traffic signals is fundamental and critical to alleviate traffic congestion in urban areas. Because of the high complexity of modelling the problem, experimental settings of current works are often inconsistent. We propose a novel and strong baseline model based on deep reinforcement learning with the encoder-decoder structure.
  arXiv  Detail & Related papers  (2021-01-24T03:55:39Z)
• Model-Based Deep Learning [155.063817656602]
  Signal processing, communications, and control have traditionally relied on classical statistical modeling techniques. Deep neural networks (DNNs) use generic architectures which learn to operate from data, and demonstrate excellent performance. We are interested in hybrid techniques that combine principled mathematical models with data-driven systems to benefit from the advantages of both approaches.
  arXiv  Detail & Related papers  (2020-12-15T16:29:49Z)
• Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks [91.65637773358347]
  We propose a general graph neural network framework designed specifically for multivariate time series data. Our approach automatically extracts the uni-directed relations among variables through a graph learning module. Our proposed model outperforms the state-of-the-art baseline methods on 3 of 4 benchmark datasets.
  arXiv  Detail & Related papers  (2020-05-24T04:02:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.