Results:
-
$f_1(x)$ over validation function$t_1(x)$ with scipyoptimize.fmin
.
(see nbviewer in Repo Visualization 2-3)
-
Maximization arg max. in LaTeX
-
Inverse Transform Method in LaTeX.
-
Inverse Transform Method with
matplotlib
.
- Acceptance-Rejection Method within
$f_1(x)$ over$t_1(x)$ .
Rejected values $\approx$ 0.256
- Acceptance-Rejection Method of
$f_1(x)$ over$t_2(x)$ .
Rejected values $\approx$ 0.486
- Acceptance-Rejection Method of
$f_2(x)$ over$t_1(x)$ and$f_2(x)$ over$t_3(x)$ .
Rejected values $\approx$ 0.446
Rejected values $\approx$ 0.506
-
Rejection sampling can be used with given/unknown functions
$f(x)$ with its densities modelled by KDEs or other methods.
-
Proposing
$t(x)$ can be hard.