Empirical sweeps on CIFAR show larger training sets improve generalization reliably while model complexity changes do not, color removal hurts performance, and added edge or wavelet features produce architecture-dependent effects.
Incorporat- ing image gradients as secondary input associated with input image to improve the performance of the cnn model.arXiv preprint arXiv:2006.04570, 2020
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
An Empirical Study of Data Scale, Model Complexity, and Input Modalities in Visual Generalization
Empirical sweeps on CIFAR show larger training sets improve generalization reliably while model complexity changes do not, color removal hurts performance, and added edge or wavelet features produce architecture-dependent effects.