Unsupervised Feature Learning through Divergent Discriminative Feature Accumulation

TL;DR

This paper proposes a novel unsupervised feature learning method called divergent discriminative feature accumulation (DDFA), which continually accumulates discriminative features without relying on reconstruction or error minimization.

Contribution

It introduces a divergence-based approach to unsupervised feature learning that accumulates features making novel discriminations, differing from traditional reconstruction-based methods.

Findings

Demonstrates the effectiveness of DDFA features on MNIST dataset

Shows DDFA continues to add useful features indefinitely

Confirms DDFA as a viable unsupervised feature learning technique

Abstract

Unlike unsupervised approaches such as autoencoders that learn to reconstruct their inputs, this paper introduces an alternative approach to unsupervised feature learning called divergent discriminative feature accumulation (DDFA) that instead continually accumulates features that make novel discriminations among the training set. Thus DDFA features are inherently discriminative from the start even though they are trained without knowledge of the ultimate classification problem. Interestingly, DDFA also continues to add new features indefinitely (so it does not depend on a hidden layer size), is not based on minimizing error, and is inherently divergent instead of convergent, thereby providing a unique direction of research for unsupervised feature learning. In this paper the quality of its learned features is demonstrated on the MNIST dataset, where its performance confirms that indeed DDFA is a viable technique for learning useful features.