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

Multitask learning (MTL) aims to develop a unified model that can handle aset of closely related tasks simultaneously. By optimizing the model acrossmultiple tasks, MTL generally surpasses its non-MTL counterparts in terms ofgeneralizability. Although MTL has been extensively researched in variousdomains such as computer vision, natural language processing, andrecommendation systems, its application to time series data has receivedlimited attention. In this paper, we investigate the application of MTL to thetime series classification (TSC) problem. However, when we integrate thestate-of-the-art 1D convolution-based TSC model with MTL, the performance ofthe TSC model actually deteriorates. By comparing the 1D convolution-basedmodels with the Dynamic Time Warping (DTW) distance function, it appears thatthe underwhelming results stem from the limited expressive power of the 1Dconvolutional layers. To overcome this challenge, we propose a novel design fora 2D convolution-based model that enhances the model's expressiveness.Leveraging this advantage, our proposed method outperforms competing approacheson both the UCR Archive and an industrial transaction TSC dataset.

Multitask Learning for Time Series Data with 2D Convolution