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

Deep learning model (primarily convolutional networks and LSTM) for timeseries classification has been studied broadly by the community with the wideapplications in different domains like healthcare, finance, industrialengineering and IoT. Meanwhile, Transformer Networks recently achieved frontierperformance on various natural language processing and computer vision tasks.In this work, we explored a simple extension of the current TransformerNetworks with gating, named Gated Transformer Networks (GTN) for themultivariate time series classification problem. With the gating that mergestwo towers of Transformer which model the channel-wise and step-wisecorrelations respectively, we show how GTN is naturally and effectivelysuitable for the multivariate time series classification task. We conductcomprehensive experiments on thirteen dataset with full ablation study. Ourresults show that GTN is able to achieve competing results with currentstate-of-the-art deep learning models. We also explored the attention map forthe natural interpretability of GTN on time series modeling. Our preliminaryresults provide a strong baseline for the Transformer Networks on multivariatetime series classification task and grounds the foundation for future research.

Gated Transformer Networks for Multivariate Time Series Classification