计算机科学 ›› 2024, Vol. 51 ›› Issue (5): 27-34.doi: 10.11896/jsjkx.230100086

• 数据库&大数据&数据科学 • 上一篇    下一篇

ST-WaveMLP:面向交通流量预测的时空全局感知网络模型

包锴楠1,2,3, 张钧波1,2,3, 宋礼2,3, 李天瑞1   

  1. 1 西南交通大学计算机与人工智能学院 成都 611756
    2 京东城市(北京)数字科技有限公司 北京 100176
    3 北京京东智能城市大数据研究院 北京 100176
  • 收稿日期:2023-01-16 修回日期:2023-06-20 出版日期:2024-05-15 发布日期:2024-05-08
  • 通讯作者: 张钧波(msjunbozhang@outlook.com)
  • 作者简介:(baokainan123@gmail.com)
  • 基金资助:
    国家重点研发计划(2019YFB2103201);国家自然科学基金(62172034);北京市科技新星(Z201100006820053)

ST-WaveMLP:Spatio-Temporal Global-aware Network for Traffic Flow Prediction

BAO Kainan1,2,3, ZHANG Junbo1,2,3, SONG Li2,3, LI Tianrui1   

  1. 1 School of Computing and Artificial Intelligence,Southwest Jiaotong University,Chengdu 611756,China
    2 JD iCity,JD Technology,Beijing 100176,China
    3 JD Intelligent Cities Research,Beijing 100176,China
  • Received:2023-01-16 Revised:2023-06-20 Online:2024-05-15 Published:2024-05-08
  • About author:BAO Kainan,born in 1996,postgra-duate.His main research interests include spatio-temporal data mining and urban computing.
    ZHANG Junbo,born in 1986,Ph.D,is a senior member of CCF(No.21344S).His main research interests include spatio-temporal data mining,urban computing,and federated learning.
  • Supported by:
    National Key R & D Program of China(2019YFB2103201), National Natural Science Foundation of China(62172034)and Beijing Nova program(Z201100006820053).

摘要: 交通流量预测在智能交通系统中起着至关重要的作用。精准的交通流量预测不仅能帮助城市管理者进行更好的交通管理,也能帮助人们制定合适的出行计划。然而精准预测交通流量颇具挑战性,主要难点在于如何捕获交通流量数据中复杂的时空依赖性。近年来,深度学习方法已被成功应用于网格交通流量预测,主要采用深度卷积神经网络来捕获时空依赖性。但是卷积神经网络主要关注数据中空间特征的提取与整合,难以充分挖掘其中复杂的时空依赖性,而且单层卷积网络只能捕获局部空间依赖,因此,要想捕获全局空间依赖就需要对超多层的卷积网络进行堆叠,这将使整个网络模型训练收敛速度变慢。为了解决些问题,提出了一种面向交通流量预测的全局感知时空网络模型ST-WaveMLP,主要使用以多层感知机(MLP)为基础的可重复结构ST-WaveBlock来捕获相关的时空依赖。ST-WaveBlock中包含了捕获全局空间依赖和局部时间依赖的模块(SGAC),以及用于捕获局部空间依赖和全局时间依赖的模块(SLAC)。ST-WaveBlock具有较强的时空表征学习能力,通常仅用2~4个ST-WaveBlock堆叠就能有效捕获数据中的时空依赖性。最后,在4个实际交通流量数据集上进行实验验证,结果表明ST-WaveMLP具有更好的收敛性以及更高的预测精度,相较于之前最好的方法,所提方法预测精度的提升最高可达9.57%,模型收敛速度的提升最高可达30.6%。

关键词: 交通流量预测, 时空依赖性, 时空深度学习, 时空数据挖掘

Abstract: Traffic flow prediction plays an incredibly important role in intelligent transportation systems.Accurate traffic flow prediction can not only benefit transportation management but also provide appropriate travel plans for people.However,it is very challenging and the main difficulty lies in how to capture the complex spatial and temporal dependencies.In recent years,deep learning methods,mainly based on convolutional neural networks,have been successfully applied to traffic forecasting tasks.However,convolutional neural networks mainly focus on the extraction and integration of spatial features in data,so it is difficult to fully explore the complex spatio-temporal dependencies.Moreover,single-layer convolutional networks can only capture the local spatial dependencies,it is necessary to stack multiple layers of convolutional networks to capture the global spatial dependencies,which will slow down the convergence speed of the whole network model training.To solve these problems,a global-aware spatio-temporal network model(called ST-WaveMLP) for traffic prediction is proposed,which mainly employs a multi-layer perceptron based repeatable structure ST-WaveBlock to capture the complex spatio-temporal dependencies.ST-WaveBlock has an excellent spatio-temporal representation learning capability,often using only 2~4 ST-WaveBlock stacks to effectively capture the spatio-temporal dependencies in the data.Finally,the experimental validation on four real traffic flow datasets shows that ST-WaveMLP has better convergence and better prediction accuracy,with a relative improvement of up to 9.57% in prediction accuracy and up to 30.6% in model convergence speed compared to the previous best method.

Key words: Traffic flow prediction, Spatio-Temporal dependencies, Spatio-Temporal deep learning, Spatio-Temporal data mining

中图分类号: 

  • TP183
[1]AGARWAL P K,GURJAR J,AGARWAL A K,et al.Application of artificial intelligence for development of intelligent transport system in smart cities[J].International Journal of Transportation Engineering and Traffic System,2015,1(2):20-30.
[2]FAN Z,SONG X,SHIBASAKI R,et al.Citymomen-tum:an online approach for crowd behavior prediction at a citywide level[C]//Proceedings of the 2015 ACM International Joint Confe-rence on Pervasive and Ubiquitous Computing.2015:559-569.
[3]ABADI A,RAJABIOUN T,IOANNOU P A.Traffic flow prediction for road transportation networks with limited traffic data[J].IEEE Transactions on Intelligent Transportation Systems,2014,16(2):653-662.
[4]HINTON G E,SALAKHUTDINOV R R.Reducing the dimensionality of data with neural networks[J].Science,2006,313(5786):504-507.
[5]ZHANG J,ZHENG Y,QI D,et al.DNN-based prediction model for spatio-temporal data[C]//Proceedings of the 24th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.2016:1-4.
[6]ZHANG J,ZHENG Y,QI D.Deep spatio-temporal residual networks for citywide crowd flows prediction[C]//Thirty-first AAAI Conference on Artificial Intelligence.2017.
[7]LIN Z,FENG J,LU Z,et al.Deepstn+:Context-aware spatial-temporal neural network for crowd flow prediction in metropolis[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2019,33(1):1020-1027.
[8]GUO S,LIN Y,LI S,et al.Deep spatial-temporal 3D convolutional neural networks for traffic data forecasting[J].IEEE Transactions on Intelligent Transportation Systems,2019,20(10):3913-3926.
[9]LI T,ZHANG J,BAO K,et al.Autost:Efficient neural architecture search for spatio-temporal prediction[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining.2020:794-802.
[10]GAO H L,XU Y B,HOU D Z,et al.Short-term traffic flow prediction algorithm for road network based on deep asynchronous residual network[J].Journal of Jilin University(Engineering and Technology Edition),2023,53(12):3458-3464.
[11]LECUN Y,BOSER B,DENKER J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural computation,1989,1(4):541-551.
[12]ROSENBLATT F.Principles of neurodynamics.perceptrons and the theory of brain mechanisms[R].Cornell Aeronautical Lab Inc Buffalo NY,1961.
[13]TOUVRON H,BOJANOWSKI P,CARON M,et al.Resmlp:Feedforward networks for image classification with data-efficient training[J].arXiv:2105.03404,2022.
[14]TOLSTIKHIN I O,HOULSBY N,KOLESNIKOV A,et al.Mlp-mixer:An all-mlp architecture for vision[J].Advances in Neural Information Processing Systems,2021,34:24261-24272.
[15]DING X,XIA C,ZHANG X,et al.Repmlp:Re-parameterizing convolutions into fully-connected layers for image recognition[J].arXiv:2105.01883,2021.
[16]TANG Y,HAN K,GUO J,et al.An image patch is a wave:Phase-aware vision mlp[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:10935-10944.
[17]TRAN D,BOURDEV L,FERGUS R,et al.Learning spatiotemporal features with 3d convolutional networks[C]//Proceedings of the IEEE International Conference on Computer Vision.2015:4489-4497.
[18]LIU H,SIMONYAN K,YANG Y.Darts:Differentiable architecture search[J].arXiv:1806.09055,2018.
[19]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.Animage is worth 16x16 words:Transformers for image recognition at scale[J].arXiv:2010.11929,2020.
[20]LIU Z,LIN Y,CAO Y,et al.Swin transformer:Hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:10012-10022.
[21]YUAN K,GUO S,LIU Z,et al.Incorporating convolution designs into visual transformers[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:579-588.
[22]SHI W,CABALLERO J,HUSZÁR F,et al.Real-time singleimage and video super-resolution using an efficient sub-pixel convolutional neural network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:1874-1883.
[23]SHI X,CHEN Z,WANG H,et al.Convolutional LSTM network:A machine learning approach for precipitation nowcasting[C]//Proceedings of the 28th International Conference on Neural Information Processing Systems(Volume 1).2015:802-810.
[24]NGUYEN Q,HEIN M.The loss surface of deep and wide neural networks[C]//International Conference on Machine Lear-ning.PMLR,2017:2603-2612.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!