Stochastic Optimization of Plain Convolutional Neural Networks with Simple methods
Yahia Assiri
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/junaidaliop/MNIST-SOPCNNpytorch★ 9
Abstract
Convolutional neural networks have been achieving the best possible accuracies in many visual pattern classification problems. However, due to the model capacity required to capture such representations, they are often oversensitive to overfitting and therefore require proper regularization to generalize well. In this paper, we present a combination of regularization techniques which work together to get better performance, we built plain CNNs, and then we used data augmentation, dropout and customized early stopping function, we tested and evaluated these techniques by applying models on five famous datasets, MNIST, CIFAR10, CIFAR100, SVHN, STL10, and we achieved three state-of-the-art-of (MNIST, SVHN, STL10) and very high-Accuracy on the other two datasets.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| CIFAR-10 | Stochastic Optimization of Plain Convolutional Neural Networks with Simple methods | Percentage correct | 94.29 | — | Unverified |
| CIFAR-100 | SOPCNN | Percentage correct | 72.96 | — | Unverified |
| MNIST | SOPCNN (Only a single Model) | Percentage error | 0.17 | — | Unverified |
| STL-10 | SOPCNN | Percentage correct | 88.08 | — | Unverified |
| SVHN | SOPCNN | Percentage error | 1.5 | — | Unverified |