Weighted Information Theoretic Models and Their Applications for Manipulating an Approximate Probability Distribution

Abstract

            The hypothetical perception about the maximum entropy principle signals that it contributes with a momentous accountability for the learning of numerous optimization problems interconnected with the information theoretic models. On the other hand, the concept of weighted information has been proved to be exceptionally productive because of its connotation in goal oriented experiments. The present communication is a step in the direction of developing two new weighted models in the probability spaces and making the study of this principle in approximating a probability distribution. With the help of these weighted models, we have demonstrated the maximum entropy principle through numerical illustration.