0 references
Controlling Covariate Shift using Balanced Normalization of Weights
arXiv (Cornell University)2018
Abstract
We introduce a new normalization technique that exhibits the fast convergence properties of batch normalization using a transformation of layer weights instead of layer outputs. The proposed technique keeps the contribution of positive and negative weights to the layer output balanced. We validate our method on a set of standard benchmarks including CIFAR-10/100, SVHN and ILSVRC 2012 ImageNet.
Related Papers
- → An introduction to the full random effects model(2022)21 cited
- → Sparse classification with paired covariates(2019)7 cited
- → Covariate Adjustment Strategy Increases Power in the Randomized Controlled Trial With Discrete-Time Survival Endpoints(2012)6 cited
- Survival analysis with coarsely observed covariates(2003)
- → Adjustment when Covariates are Fallible(2016)6 cited