APAC: Augmented PAttern Classification with Neural Networks

TL;DR

This paper introduces APAC, a novel classification method that optimally utilizes augmented data, significantly improving generalization and achieving state-of-the-art results on benchmark datasets.

Contribution

The paper proposes APAC, an optimal decision rule for classifiers trained with augmented data, enhancing generalization performance over traditional classification methods.

Findings

APAC outperforms traditional classifiers in generalization.

CNN with APAC achieves state-of-the-art on MNIST.

MLP with APAC surpasses some recent regularization techniques on CIFAR-10.

Abstract

Deep neural networks have been exhibiting splendid accuracies in many of visual pattern classification problems. Many of the state-of-the-art methods employ a technique known as data augmentation at the training stage. This paper addresses an issue of decision rule for classifiers trained with augmented data. Our method is named as APAC: the Augmented PAttern Classification, which is a way of classification using the optimal decision rule for augmented data learning. Discussion of methods of data augmentation is not our primary focus. We show clear evidences that APAC gives far better generalization performance than the traditional way of class prediction in several experiments. Our convolutional neural network model with APAC achieved a state-of-the-art accuracy on the MNIST dataset among non-ensemble classifiers. Even our multilayer perceptron model beats some of the convolutional models with recently invented stochastic regularization techniques on the CIFAR-10 dataset.