On the Inverse Problem of Entropy Maximizations