Figure 5.6: This diagrams shows the minimum KL divergence solution between a Gaussian and a mixture of two Gaussians. The first Gaussian has a covariance matrix with a very high correlation, its eigenvalues are 1 and 0.01 and the eigenvectors are at forty-five degrees to the axis. Three contours from this Gaussian are shown with solid lines. The Gaussians in the mixture model are constrained to have diagonal covariance. Three contours from the mixture model are shown with dotted lines.