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

Add code
Aug 17, 2021
Jacek Klimek, Jakub Klimek, Witold Kraskiewicz, Mateusz Topolewski
Figure 1 for Long-term series forecasting with Query Selector -- efficient model of sparse attention
Figure 2 for Long-term series forecasting with Query Selector -- efficient model of sparse attention
Figure 3 for Long-term series forecasting with Query Selector -- efficient model of sparse attention
Figure 4 for Long-term series forecasting with Query Selector -- efficient model of sparse attention

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

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, Helpdesk and BPI'12 datasets.

* 10 pages  
View paper onarxiv icon

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon