Difference between revisions of "Selection"

From Adaptive Population based Simplex
Jump to: navigation, search
m
m (Adaptive probability)
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
== Basic selection (s0) ==
 
== Basic selection (s0) ==
# select a simplex S (i.e. $D+1$ individuals) at random;
+
# Select a simplex $S$ (i.e. $D+1$ individuals) at random. Be sure that the current one is in this list. If not, replace the last of the list by the current one.
# select the three first ones;
+
# Select the three first ones.
# sort them, by increasing order of value (fitness). They are $x_{best}$, $x_{worst2}$, and $w_{worst}$.  
+
# Sort them, by increasing order of value (fitness). They are $x_{best}$, $x_{worst2}$, and $w_{worst}$.  
  
 
If $x_{best}$ is better than the best ever found (i.e. $Best$), set $Best=x_{best}$.
 
If $x_{best}$ is better than the best ever found (i.e. $Best$), set $Best=x_{best}$.
 +
 +
=== Adaptive probability ===
 +
The rationale is the following:<br />
 +
the more the simplex volume increases, the less the algorithm is successful, and the more one needs randomness to increase the diversity.
 +
 +
Compute the volume $V(1)$ of the simplex. If the previous volume is zero (in practice, too small) the probability $p$ is set to 0.5. If not, the formula is
 +
$$
 +
p=\frac{1}{1+e^{-\frac{V(1)-V(0)}{V(0)}}}
 +
$$
 +
 +
After that, we simply set $V(0)=V(1)$, to save the simplex volume as the "previous" one.

Latest revision as of 17:15, 28 June 2013

Basic selection (s0)

  1. Select a simplex $S$ (i.e. $D+1$ individuals) at random. Be sure that the current one is in this list. If not, replace the last of the list by the current one.
  2. Select the three first ones.
  3. Sort them, by increasing order of value (fitness). They are $x_{best}$, $x_{worst2}$, and $w_{worst}$.

If $x_{best}$ is better than the best ever found (i.e. $Best$), set $Best=x_{best}$.

Adaptive probability

The rationale is the following:
the more the simplex volume increases, the less the algorithm is successful, and the more one needs randomness to increase the diversity.

Compute the volume $V(1)$ of the simplex. If the previous volume is zero (in practice, too small) the probability $p$ is set to 0.5. If not, the formula is $$ p=\frac{1}{1+e^{-\frac{V(1)-V(0)}{V(0)}}} $$

After that, we simply set $V(0)=V(1)$, to save the simplex volume as the "previous" one.