Using Early-Stopping to Avoid Overfitting in Wrapper-Based Feature Selection Employing Stochastic Search

Abstract

It is acknowledged that overfitting can occur in feature selection using the wrapper method when there is a limited amount of training data available. It has also been shown that the severity of overfitting is related to the intensity of the search algorithm used during this process. In this paper we show that two stochastic search techniques (Simulated Annealing and Genetic Algorithms) that can be used for wrapper-based feature selection are susceptible to overfitting in this way. However, because of their stochastic nature, these algorithms can be stopped early to prevent overfitting. We present a framework that implements early-stopping for both of these stochastic search techniques and we show that this is successful in reducing the effects of overfitting and in increasing generalisation accuracy in most cases.