Issue of over fitting with Wide ResNet

vision
1

I’m trying to build a Wide ResNet (WRN - 16 - 4) on CIFAR10 from scratch. I encounter over fitting problem with training accuracy at 91% and validation accuracy at 25%. The following are measures I tried but couldn’t get a solution,

  1. Drop out - 40%
  2. Data augmentation and normalization
  3. Xavier Uniform weight initialization
  4. Batch size

Kindly help me with ideas I can try to solve this issue.