In this exercise, we explore the behaviour of the Gaussian d

Question and Solution

In this exercise, we explore the behaviour of the Gaussian distribution in high-dimensional spaces. Consider a Gaussian

91 % (677 Review)

In this exercise, we explore the behaviour of the Gaussian distribution in high-dimensional spaces. Consider a Gaussian distribution in D dimensions given by p(x) = 1 (2??2)D/2 exp ?x2 2?2 . (1.147) We wish to find the density with respect to radius in polar coordinates in which the direction variables have been integrated out. To do this, show that the integral of the probability density over a thin shell of radius r and thickness , where 1, is given by p(r) where p(r) = SDrD?1 (2??2)D/2 exp ? r2 2?2 (1.148) where SD is the surface area of a unit sphere in D dimensions. Show that the function p(r) has a single stationary point located, for large D, at r ? D?. By considering p(r + ) where r, show that for large D, p(r + ) = p(r) exp ? 3 2 2?2 (1.149) which shows that r is a maximum of the radial probability density and also that p(r) decays exponentially away from its maximum at r with length scale ?. We have already seen that ? r for large D, and so we see that most of the probability mass is concentrated in a thin shell at large radius. Finally, show that the probability density p(x) is larger at the origin than at the radius r by a factor of exp(D/2). We therefore see that most of the probability mass in a high-dimensional Gaussian distribution is located at a different radius from the region of high probability density. This property of distributions in spaces of high dimensionality will have important consequences when we consider Bayesian inference of model parameters in later chapters

Copyright@2020 Crazy Prep Pvt. Ltd. (Crazy For Study)

Secured By:

Disclaimer: Crazy For Study provides academic assistance to students so that they can complete their college assignments and projects on time. We strictly do not deliver the reference papers. This is just to make you understand and used for the analysis and reference purposes only.