SGA replaces query-key projections in self-attention with a shared learnable matrix and residual term to achieve linear complexity in look-back length for time series forecasting.
The capacity and robustness trade-off: Revisiting the channel independent strategy for multivariate time series forecasting,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Self-Gating Attention for Efficient Time Series Forecasting
SGA replaces query-key projections in self-attention with a shared learnable matrix and residual term to achieve linear complexity in look-back length for time series forecasting.