A Multi-View Multi-Task Learning Framework for Multi-Variate Time Series Forecasting

Deng，Jinliang1; Chen，Xiusi2; Jiang，Renhe3; Song，Xuan4; Tsang，Ivor W.1

2022

10.1109/TKDE.2022.3218803

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING   Impact Factor & Quartile

1041-4347

1558-2191

Multi-variate time series (MTS) data is a ubiquitous class of data abstraction in the real world. Any instance of MTS is generated from a hybrid dynamical system and their specific dynamics are usually unknown. The hybrid nature of such a dynamical system is a result of complex external attributes, such as geographic location and time of day, each of which can be categorized into either spatial attributes or temporal attributes. Therefore, there are two fundamental views which can be used to analyze MTS data, namely the spatial view and the temporal view. Moreover, from each of these two views, we can partition the set of data samples of MTS into disjoint forecasting tasks in accordance with their associated attribute values. Then, samples of the same task will manifest similar forthcoming pattern, which is less sophisticated to be predicted in comparison with the original single-view setting. Considering this insight, we propose a novel multi-view multi-task (MVMT) learning framework for MTS forecasting. Instead of being explicitly presented in most scenarios, MVMT information is deeply concealed in the MTS data, which severely hinders the model from capturing it naturally. To this end, we develop two kinds of basic operations, namely task-wise affine transformation and task-wise normalization, respectively. Applying these two operations with prior knowledge on the spatial and temporal view allows the model to adaptively extract MVMT information while predicting. Extensive experiments on three datasets are conducted to illustrate that canonical architectures can be greatly enhanced by the MVMT learning framework in terms of both effectiveness and efficiency. In addition, we design rich case studies to reveal the properties of representations produced at different phases in the entire prediction procedure.

Correlation Deep learning Forecasting multi-view multi-task learning Multitasking normalization Predictive models Task analysis Time series analysis time series forecasting Transformers

English

Others

ENGINEERING

2-s2.0-85141598882

Scopus

1.Australian Artificial Intelligence Institute, University of Technology Sydney, Sydney, Australia
2.University of California, Los Angeles, USA
3.Center for Spatial Information Science, University of Tokyo, Tokyo, Japan
4.SUSTech-UTokyo Joint Research Center on Super Smart City, Department of Computer Science and Engineering, Southern University of Science and Technology (SUSTech), Shenzhen, China

Deng，Jinliang,Chen，Xiusi,Jiang，Renhe,et al. A Multi-View Multi-Task Learning Framework for Multi-Variate Time Series Forecasting[J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,2022,PP(99):1-16.

Deng，Jinliang,Chen，Xiusi,Jiang，Renhe,Song，Xuan,&Tsang，Ivor W..(2022).A Multi-View Multi-Task Learning Framework for Multi-Variate Time Series Forecasting.IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,PP(99),1-16.

Deng，Jinliang,et al."A Multi-View Multi-Task Learning Framework for Multi-Variate Time Series Forecasting".IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING PP.99(2022):1-16.