Long-term series forecasting with Query Selector

07/28/2021

∙

Various modifications of TRANSFORMER were recently used to solve time-series forecasting problem. We propose Query Selector - an efficient, deterministic algorithm for sparse attention matrix. Experiments show it achieves state-of-the art results on ETT data set.

READ FULL TEXT

Long-term series forecasting with Query Selector

Long-term series forecasting with Query Selector – efficient model of sparse attention

Are Transformers Effective for Time Series Forecasting?

TDSTF: Transformer-based Diffusion probabilistic model for Sparse Time series Forecasting

MPR-Net:Multi-Scale Pattern Reproduction Guided Universality Time Series Interpretable Forecasting

A Differential Attention Fusion Model Based on Transformer for Time Series Forecasting

Does Long-Term Series Forecasting Need Complex Attention and Extra Long Inputs?

Aggregating Strategies for Long-term Forecasting

Long-term series forecasting with Query Selector

Related Research

Long-term series forecasting with Query Selector – efficient model of sparse attention

Are Transformers Effective for Time Series Forecasting?

TDSTF: Transformer-based Diffusion probabilistic model for Sparse Time series Forecasting

MPR-Net:Multi-Scale Pattern Reproduction Guided Universality Time Series Interpretable Forecasting

A Differential Attention Fusion Model Based on Transformer for Time Series Forecasting

Does Long-Term Series Forecasting Need Complex Attention and Extra Long Inputs?

Aggregating Strategies for Long-term Forecasting