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

Irregularly sampled time series are increasingly prevalent, particularly inmedical domains. While various specialized methods have been developed tohandle these irregularities, effectively modeling their complex dynamics andpronounced sparsity remains a challenge. This paper introduces a novelperspective by converting irregularly sampled time series into line graphimages, then utilizing powerful pre-trained vision transformers for time seriesclassification in the same way as image classification. This method not onlylargely simplifies specialized algorithm designs but also presents thepotential to serve as a universal framework for time series modeling.Remarkably, despite its simplicity, our approach outperforms state-of-the-artspecialized algorithms on several popular healthcare and human activitydatasets. Especially in the rigorous leave-sensors-out setting where a portionof variables is omitted during testing, our method exhibits strong robustnessagainst varying degrees of missing observations, achieving an impressiveimprovement of 42.8% in absolute F1 score points over leading specializedbaselines even with half the variables masked. Code and data are available athttps://github.com/Leezekun/ViTST

Time Series as Images: Vision Transformer for Irregularly Sampled Time Series