AIMA Artificial Intelligence a modern approch

AIMA-exercises is an open-source community of students, instructors and developers. Anyone can add an exercise, suggest answers to existing questions, or simply help us improve the platform. We accept contributions on this github repository.

Exercise 20.8 [beta-integration-exercise]

This exercise investigates properties of the Beta distribution defined in Equation (beta-equation).

  1. By integrating over the range $[0,1]$, show that the normalization constant for the distribution $[a,b]$ is given by $\alpha = \Gamma(a+b)/\Gamma(a)\Gamma(b)$ where $\Gamma(x)$ is the Gamma function, defined by $\Gamma(x+1)x\cdot\Gamma(x)$ and $\Gamma(1)1$. (For integer $x$, $\Gamma(x+1)x!$.)

  2. Show that the mean is $a/(a+b)$.

  3. Find the mode(s) (the most likely value(s) of $\theta$).

  4. Describe the distribution $[\epsilon,\epsilon]$ for very small $\epsilon$. What happens as such a distribution is updated?

View Answer