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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this work, we focus on robust time series representation learning. Ourassumption is that real-world time series is noisy and complementaryinformation from different views of the same time series plays an importantrole while analyzing noisy input. Based on this, we create two views for theinput time series through two different encoders. We conduct co-training basedcontrastive learning iteratively to learn the encoders. Our experimentsdemonstrate that this co-training approach leads to a significant improvementin performance. Especially, by leveraging the complementary information fromdifferent views, our proposed TS-CoT method can mitigate the impact of datanoise and corruption. Empirical evaluations on four time series benchmarks inunsupervised and semi-supervised settings reveal that TS-CoT outperformsexisting methods. Furthermore, the representations learned by TS-CoT cantransfer well to downstream tasks through fine-tuning.

A Co-training Approach for Noisy Time Series Learning