KL Divergence: Measuring Distributional Difference - Information Theory for Machine Learning | OneNoughtOne