Anonymous user
Unbias a random generator: Difference between revisions
Added D version
(Clarify how to unbias using mathematical properties, rm draft status (no problems for a long time)) |
(Added D version) |
||
Line 7:
The actual unbiasing should be done by generating two numbers at a time from randN and only returning a 1 or 0 if they are different. As long as you always return the first number or always return the second number, the probabilities discussed above should take over the biased probability of randN.
=={{header|
<lang d>import std.stdio, std.random;
bool biased(int n) {
return uniform(0.0, 1.0) < (1.0 / n);
}
bool unbiased(int n) {
bool a, b;
do {
a = biased(n);
b = biased(n);
} while (a == b);
return a;
}
void main() {
int M = 50_000;
foreach (n; 3 .. 7) {
int c1, c2;
foreach (i; 0 .. M) {
c1 += biased(n);
c2 += unbiased(n);
}
writefln("%d: %2.2f%% %2.2f%%", n, 100.0*c1/M, 100.0*c2/M);
}
}</lang>
Output:
<pre>3: 33.12% 50.15%
4: 25.13% 50.02%
5: 19.61% 50.01%
6: 16.70% 49.98%</pre>
=={{header|J}}==
<lang j>randN=: 0 = ?
unbiased=: i.@# { ::$: 2 | 0 3 -.~ _2 #.\ 4&* randN@# ]</lang>
| |||