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

The past decade has witnessed significant advances in time series modelingwith deep learning. While achieving state-of-the-art results, thebest-performing architectures vary highly across applications and domains.Meanwhile, for natural language processing, the Generative Pre-trainedTransformer (GPT) has demonstrated impressive performance via training onegeneral-purpose model across various textual datasets. It is intriguing toexplore whether GPT-type architectures can be effective for time series,capturing the intrinsic dynamic attributes and leading to significant accuracyimprovements. In this paper, we propose a novel framework, TEMPO, that caneffectively learn time series representations. We focus on utilizing twoessential inductive biases of the time series task for pre-trained models: (i)decomposition of the complex interaction between trend, seasonal and residualcomponents; and (ii) introducing the selection-based prompts to facilitatedistribution adaptation in non-stationary time series. TEMPO expands thecapability for dynamically modeling real-world temporal phenomena from datawithin diverse domains. Our experiments demonstrate the superior performance ofTEMPO over state-of-the-art methods on a number of time series benchmarkdatasets. This performance gain is observed not only in standard supervisedlearning settings but also in scenarios involving previously unseen datasets aswell as in scenarios with multi-modal inputs. This compelling findinghighlights TEMPO's potential to constitute a foundational model-buildingframework.

TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting