Discriminative Autoencoder for Feature Extraction: Application to Character Recognition

TL;DR

This paper introduces a discriminative autoencoder that leverages supervised learning to produce robust feature representations for character recognition, outperforming existing deep architectures in accuracy.

Contribution

The paper proposes a novel discriminative autoencoder architecture that enhances feature extraction for image classification, demonstrating superior performance over existing models.

Findings

High classification accuracy with simple classifiers like KNN

Superior representation learning compared to existing autoencoders

Effective on standard character/image recognition datasets

Abstract

Conventionally, autoencoders are unsupervised representation learning tools. In this work, we propose a novel discriminative autoencoder. Use of supervised discriminative learning ensures that the learned representation is robust to variations commonly encountered in image datasets. Using the basic discriminating autoencoder as a unit, we build a stacked architecture aimed at extracting relevant representation from the training data. The efficiency of our feature extraction algorithm ensures a high classification accuracy with even simple classification schemes like KNN (K-nearest neighbor). We demonstrate the superiority of our model for representation learning by conducting experiments on standard datasets for character/image recognition and subsequent comparison with existing supervised deep architectures like class sparse stacked autoencoder and discriminative deep belief network.