4.7 Article

Spatio-Temporal Meta Learning for Urban Traffic Prediction

Journal

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
Volume 34, Issue 3, Pages 1462-1476

Publisher

IEEE COMPUTER SOC
DOI: 10.1109/TKDE.2020.2995855

Keywords

Urban traffic; spatio-temporal data; neural network; meta learning

Funding

  1. National Key R&D Program of China [2019YFB2101805]
  2. APEX-MSRA Joint Research Program
  3. Natural Science Foundation of China [61672399, U1609217, 61773324, 61702327, 61772333, 61632017]

Ask authors/readers for more resources

This paper proposes a deep meta learning model, ST-MetaNet(+), for predicting urban traffic. The model captures complex spatio-temporal correlations using an encoder and a decoder, and generates weights using embeddings of geo-graph attributes and traffic context. Experimental results demonstrate the effectiveness of ST-MetaNet(+) in surpassing other methods.
Predicting urban traffic is of great importance to intelligent transportation systems and public safety, yet is very challenging in three aspects: 1) complex spatio-temporal correlations of urban traffic, including spatial correlations between locations along with temporal correlations among timestamps; 2) spatial diversity of such spatio-temporal correlations, which varies from location to location and depends on the surrounding geographical information, e.g., points of interests and road networks; and 3) temporal diversity of such spatio-temporal correlations, which is highly influenced by dynamic traffic states. To tackle these challenges, we proposed a deep meta learning based model, entitled ST-MetaNet(+), to collectively predict traffic in all locations at the same time. ST-MetaNet(+) employs a sequence-to-sequence architecture, consisting of an encoder to learn historical information and a decoder to make predictions step by step. Specifically, the encoder and decoder have the same network structure, consisting of meta graph attention networks and meta recurrent neural networks, to capture diverse spatial and temporal correlations, respectively. Furthermore, the weights (parameters) of meta graph attention networks and meta recurrent neural networks are generated from the embeddings of geo-graph attributes and the traffic context learned from dynamic traffic states. Extensive experiments were conducted based on three real-world datasets to illustrate the effectiveness of ST-MetaNet(+) beyond several state-of-the-art methods.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available