A deep architecture combining CNNS and GRBMS for traffic speed prediction

Huachun Tan, Zhiyu Zhong, Yuankai Wu, Xiaoxuan Chen, Jian Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Short-term traffic speed prediction is an important technique for advanced traffic management systems. In this paper, a deep architecture combining convolutional neural networks (CNNs) and restricted Boltzmann machines (RBMs) with Gaussian units (GRBM-CNN) is proposed to predict short-term traffic speed; it combines the advantages of RBMs and CNNs and achieves a powerful capability to learn complex features of traffic data. Furthermore, this paper uses a graphics processing unit (GPU) to accelerate the learning process. Experimental results show that the proposed deep architecture has a lot of advantages over state-of-arts on the short-term traffic speed prediction task.

Original languageEnglish
Title of host publicationCICTP 2017
Subtitle of host publicationTransportation Reform and Change - Equity, Inclusiveness, Sharing, and Innovation - Proceedings of the 17th COTA International Conference of Transportation Professionals
EditorsHaizhong Wang, Jian Sun, Jian Lu, Lei Zhang, Yu Zhang, ShouEn Fang
PublisherAmerican Society of Civil Engineers (ASCE)
Pages310-319
Number of pages10
ISBN (Electronic)9780784480915
DOIs
Publication statusPublished - 2018
Event17th COTA International Conference of Transportation Professionals: Transportation Reform and Change - Equity, Inclusiveness, Sharing, and Innovation, CICTP 2017 - Shanghai, China
Duration: 7 Jul 20179 Jul 2017

Publication series

NameCICTP 2017: Transportation Reform and Change - Equity, Inclusiveness, Sharing, and Innovation - Proceedings of the 17th COTA International Conference of Transportation Professionals
Volume2018-January

Conference

Conference17th COTA International Conference of Transportation Professionals: Transportation Reform and Change - Equity, Inclusiveness, Sharing, and Innovation, CICTP 2017
Country/TerritoryChina
CityShanghai
Period7/07/179/07/17

Keywords

  • Convolutional Neural Networks (CNNs)
  • Deep learning
  • Restricted Boltzmann Machines (RBMs)
  • Traffic speed prediction

Fingerprint

Dive into the research topics of 'A deep architecture combining CNNS and GRBMS for traffic speed prediction'. Together they form a unique fingerprint.

Cite this