Timezone: »

 
Spotlight
AutoST: Towards the Universal Modeling of Spatio-temporal Sequences
Jianxin Li · Shuai Zhang · Hui Xiong · Haoyi Zhou

Tue Dec 06 05:00 PM -- 07:00 PM (PST) @
in Lightning Talks 2B-2 »

The analysis of spatio-temporal sequences plays an important role in many real-world applications, demanding a high model capacity to capture the interdependence among spatial and temporal dimensions. Previous studies provided separated network design in three categories: spatial first, temporal first, and spatio-temporal synchronous. However, the manually-designed heterogeneous models can hardly meet the spatio-temporal dependency capturing priority for various tasks. To address this, we proposed a universal modeling framework with three distinctive characteristics: (i) Attention-based network backbone, including S2T Layer (spatial first), T2S Layer (temporal first), and STS Layer (spatio-temporal synchronous). (ii) The universal modeling framework, named UniST, with a unified architecture that enables flexible modeling priorities with the proposed three different modules. (iii) An automatic search strategy, named AutoST, automatically searches the optimal spatio-temporal modeling priority by network architecture search. Extensive experiments on five real-world datasets demonstrate that UniST with any single type of our three proposed modules can achieve state-of-the-art performance. Furthermore, AutoST can achieve overwhelming performance with UniST.

Keywords: Neural Network · Architecture Search · Spatio-temporal Forecasting

Author Information

Jianxin Li (Beihang University)
Shuai Zhang (Beihang University)
Hui Xiong
Haoyi Zhou (Beihang University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors