Proposing a reinforcement learning based approach for feature selection

In supervised learning scenarios, feature selection has been studied widely in the literature. Here, feature selection is considered as an empirical strategy of restricting state space and lessen the complexity of hypothesis. In this work we introduce the environment as a one player game and improve a reinforcement learning method to traverse the state space and learn from experiments. In this way we exert a Monte Carlo based method, and adapt the problem with the requirements of such procedure. As we need to reward each action and evaluate the states, we employ SVM classifier according to selected features. Moreover SVM is used to determine the appropriation of selected features and evaluate the final subset of features. At last we compare this work with the state of the art methods. Experimental results demonstrate the effectiveness and efficiency of our algorithm.