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

We propose SutraNets, a novel method for neural probabilistic forecasting oflong-sequence time series. SutraNets use an autoregressive generative model tofactorize the likelihood of long sequences into products of conditionalprobabilities. When generating long sequences, most autoregressive approachessuffer from harmful error accumulation, as well as challenges in modelinglong-distance dependencies. SutraNets treat long, univariate prediction asmultivariate prediction over lower-frequency sub-series. Autoregressionproceeds across time and across sub-series in order to ensure coherentmultivariate (and, hence, high-frequency univariate) outputs. Since sub-seriescan be generated using fewer steps, SutraNets effectively reduce erroraccumulation and signal path distances. We find SutraNets to significantlyimprove forecasting accuracy over competitive alternatives on six real-worlddatasets, including when we vary the number of sub-series and scale up thedepth and width of the underlying sequence models.

SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecasting