Information Maximization Perspective of Orthogonal Matching Pursuit with Applications to Explainable AI