ReWA uses reparameterization with weight decay and adaptive rates to create a stable optimization landscape connected to ℓ_p regularization, yielding higher sparsity than ℓ1 on ResNets for CIFAR-10 and ImageNet while preserving accuracy.
More is less: Inducing sparsity via overparameterization
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
-
Theoretical Analysis of Sparse Optimization with Reparameterization, Weight Decay, and Adaptive Learning Rate
ReWA uses reparameterization with weight decay and adaptive rates to create a stable optimization landscape connected to ℓ_p regularization, yielding higher sparsity than ℓ1 on ResNets for CIFAR-10 and ImageNet while preserving accuracy.