APAC: Augmented PAttern Classification with Neural Networks

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.

References in corpus (7)

• Improving neural networks by preventing co-adaptation of feature detectors
• Going Deeper with Convolutions
• Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition
• Deeply-Supervised Nets
• Stochastic Pooling for Regularization of Deep Convolutional Neural Networks
• Deep Image: Scaling up Image Recognition
• Improving Deep Neural Networks with Probabilistic Maxout Units

Cited by in corpus (25)

• AutoAugment: Learning Augmentation Policies from Data
• FILIP: Fine-grained Interactive Language-Image Pre-Training
• Generalization in Deep Learning
• Dynamic Routing Between Capsules
• Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures
• Affinity and Diversity: Quantifying Mechanisms of Data Augmentation
• Ensemble learning in CNN augmented with fully connected subnetworks
• UniformAugment: A Search-free Probabilistic Data Augmentation Approach
• Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Data
• Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Data
• PseudoAugment: Learning to Use Unlabeled Data for Data Augmentation in Point Clouds
• RoIMix: Proposal-Fusion among Multiple Images for Underwater Object Detection
• Improving 3D Object Detection through Progressive Population Based Augmentation
• JuncNet: A Deep Neural Network for Road Junction Disambiguation for Autonomous Vehicles
• Open Domain Generalization with Domain-Augmented Meta-Learning
• Low-cost and high-performance data augmentation for deep-learning-based skin lesion classification
• Data Interpolating Prediction: Alternative Interpretation of Mixup
• CLCC: Contrastive Learning for Color Constancy
• Does Data Augmentation Benefit from Split BatchNorms
• Automatically Learning Data Augmentation Policies for Dialogue Tasks
• GCCN: Global Context Convolutional Network
• Signature-Graph Networks
• Bridging the Band Gap: What Device Physicists Need to Know About Machine Learning
• Inspect Transfer Learning Architecture with Dilated Convolution
• Global and Local Texture Randomization for Synthetic-to-Real Semantic Segmentation