By the way I think that these testcases are much more relevant for noisy
optimization than the BBOB noisy optimization,
in which noise is not the main issue. But maybe there is a lot of room for
debates around that, and it becomes far from computer-go :-)
Olivier
___
Yes. f(x) is not the output. The output is either 0 or 1, and f(x) is the
probability of 1.
Rémi
On 6 mars 2013, at 09:04, Chin-Chang Yang wrote:
> Thank you, Olivier.
>
> Let the observable function value be o(x). It can be defined as:
>
> o(x) = 1, with probability f(x);
> o(x) = 0, with
Thank you, Olivier.
Let the observable function value be o(x). It can be defined as:
o(x) = 1, with probability f(x);
o(x) = 0, with probability (1 - f(x)).
where f(x) = 1 / (1 + e(-r(x))) has been defined in the paper. Also, we can
see that the expected value is f(x).
Did I get this correct?
B
