Efficient Wrapper Feature Selection using Autoencoder and Model Based Elimination

[ArXiv] [PDF] [Code] [Cite]

Abstract

We propose a computationally efficient wrapper feature selection method - called Autoencoder and Model Based Elimination of features using Relevance and Redundancy scores (AMBER) - that uses a single ranker model along with autoencoders to perform greedy backward elimination of features, without requiring model retraining. The ranker model is used to prioritize the removal of features that are not critical to the classification task, while the autoencoders are used to prioritize the elimination of correlated features. We demonstrate the superior feature selection ability of AMBER on four well known datasets corresponding to different domain applications via comparing the accuracies with other computationally efficient state of the art feature selection techniques, and note how a surprisingly small number of features can lead to very high accuracies on some datasets.