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

Due to effective pattern mining and feature representation, neuralforecasting models based on deep learning have achieved great progress. Thepremise of effective learning is to collect sufficient data. However, in timeseries forecasting, it is difficult to obtain enough data, which limits theperformance of neural forecasting models. To alleviate the data scarcitylimitation, we design Sequence Decomposition Adaptation Network (SeDAN) whichis a novel transfer architecture to improve forecasting performance on thetarget domain by aligning transferable knowledge from cross-domain datasets.Rethinking the transferability of features in time series data, we proposeImplicit Contrastive Decomposition to decompose the original features intocomponents including seasonal and trend features, which are easier to transfer.Then we design the corresponding adaptation methods for decomposed features indifferent domains. Specifically, for seasonal features, we perform jointdistribution adaptation and for trend features, we design an Optimal LocalAdaptation. We conduct extensive experiments on five benchmark datasets formultivariate time series forecasting. The results demonstrate the effectivenessof our SeDAN. It can provide more efficient and stable knowledge transfer.

Improving the Transferability of Time Series Forecasting with Decomposition Adaptation