A dynamic global plus component loss weighting modestly improves some decomposition-based forecasters, but the paper overstates its results and mischaracterizes PatchTST.
Therefore, we show the results of the forecasting part with the settings of the input length 96 and prediction length{192, 336, 720} here, respectively
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
A Hybrid Loss Framework for Decomposition-based Time Series Forecasting Methods: Balancing Global and Component Errors
A dynamic global plus component loss weighting modestly improves some decomposition-based forecasters, but the paper overstates its results and mischaracterizes PatchTST.