Displaying 1 result from an estimated 1 matches for "fnrosenbrock".
2023 Aug 13
4
Noisy objective functions
...pplications, I got interested in
optimizing noisy objective functions. As an (artificial) example, the
following is the Rosenbrock function, where Gaussian noise of standard
deviation `sd = 0.01` is added to the function value.
fn <- function(x)
(1+rnorm(1, sd=0.01)) * adagio::fnRosenbrock(x)
To smooth out the noise, define another function `fnk(x, k = 1)` that
calls `fn` k times and returns the mean value of those k function
applications.
fnk <- function(x, k = 1) { # fnk(x) same as fn(x)
rv = 0.0
for (i in 1:k) rv <- rv + fn(x)
return(rv/n)...