An Ensemble of Simple Convolutional Neural Network Models for MNIST Digit Recognition

12 Aug 2020  ·  Sanghyeon An, Minjun Lee, Sanglee Park, Heerin Yang, Jungmin So ·

We report that a very high accuracy on the MNIST test set can be achieved by using simple convolutional neural network (CNN) models. We use three different models with 3x3, 5x5, and 7x7 kernel size in the convolution layers. Each model consists of a set of convolution layers followed by a single fully connected layer. Every convolution layer uses batch normalization and ReLU activation, and pooling is not used. Rotation and translation is used to augment training data, which is frequently used in most image classification tasks. A majority voting using the three models independently trained on the training data set can achieve up to 99.87% accuracy on the test set, which is one of the state-of-the-art results. A two-layer ensemble, a heterogeneous ensemble of three homogeneous ensemble networks, can achieve up to 99.91% test accuracy. The results can be reproduced by using the code at:

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Image Classification MNIST Heterogeneous ensemble with simple CNN Percentage error 0.09 # 1
Accuracy 99.91 # 1