| Author |
Comment/Response |
Euler314
|
 05/11/08 2:20pm
Hi. I am looking for built-in functions or packages that can do the following:
1. Given a function f : R^k -> R, find the point in which it attains a global maximum, without trying to calculate the derivatives of f. My function f doesn't have a nice analytical form, but rather is computed according to some algorithm. Functions like Maximize and FindMaximum seem to first try to evaluate the function and compute its gradient - I want something that will treat f as a black box.
2. Given functions f[1], ..., f[k] : R^k -> R, find a point in which a Nash equilibrium is achieved, that is, X = {x[1], ..., x[k]} such that for any 1 <= i <= k and any t, f[k][{x[1],...,x[i-1],t,x[i+1],...,x[k]] <= f[k][X]. Ideally this should also treat the functions as black boxes.
Any suggestions? Thanks.
URL: , |
|
