English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multi-variate time series (MTS) data is a ubiquitous class of dataabstraction in the real world. Any instance of MTS is generated from a hybriddynamical system and their specific dynamics are usually unknown. The hybridnature 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 categorizedinto either spatial attributes or temporal attributes. Therefore, there are twofundamental views which can be used to analyze MTS data, namely the spatialview and the temporal view. Moreover, from each of these two views, we canpartition the set of data samples of MTS into disjoint forecasting tasks inaccordance with their associated attribute values. Then, samples of the sametask will manifest similar forthcoming pattern, which is less sophisticated tobe predicted in comparison with the original single-view setting. Consideringthis insight, we propose a novel multi-view multi-task (MVMT) learningframework for MTS forecasting. Instead of being explicitly presented in mostscenarios, MVMT information is deeply concealed in the MTS data, which severelyhinders the model from capturing it naturally. To this end, we develop twokinds of basic operations, namely task-wise affine transformation and task-wisenormalization, respectively. Applying these two operations with prior knowledgeon the spatial and temporal view allows the model to adaptively extract MVMTinformation while predicting. Extensive experiments on three datasets areconducted to illustrate that canonical architectures can be greatly enhanced bythe MVMT learning framework in terms of both effectiveness and efficiency. Inaddition, we design rich case studies to reveal the properties ofrepresentations produced at different phases in the entire predictionprocedure.

A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting