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

Time series forecasting task predicts future trends based on historicalinformation. Recent U-Net-based methods have demonstrated superior performancein predicting real-world datasets. However, the performance of these models islower than patch-based models or linear models. In this work, we propose asymmetric and hierarchical framework, Kernel-U-Net, which cuts the inputsequence into slices at each layer of the network and then computes them usingkernels. Furthermore, it generalizes the concept of convolutional kernels inclassic U-Net to accept custom kernels that follow the same design pattern.Compared to the existing linear or transformer-based solution, our modelcontains 3 advantages: 1) A small number of parameters: the parameters size is$O(log(L)^2)$ where $L$ is the look-back window size, 2) Flexibility: itskernels can be customized and fitted to the datasets, 3) Computationefficiency: the computation complexity of transformer modules is reduced to$O(log(L)^2)$ if they are placed close to the latent vector. Kernel-U-Netaccuracy was greater than or equal to the state-of-the-art model on six (out ofseven) real-world datasets.

Kernel-U-Net: Hierarchical and Symmetrical Framework for Multivariate Time Series Forecasting