Adaptive Resolution Residual Networks use Laplacian residuals and Laplacian dropout to make fixed-resolution layers adaptive across image resolutions with a guaranteed computation-skipping property under ideal kernels.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Resolution Residual Networks -- Generalizing Across Resolutions Easily and Efficiently
Adaptive Resolution Residual Networks use Laplacian residuals and Laplacian dropout to make fixed-resolution layers adaptive across image resolutions with a guaranteed computation-skipping property under ideal kernels.