An Automatic Method for Finding the Greatest or Least Value of a Function
Abstract¶
The greatest or least value of a function of several variables is to be found when the variables are restricted to a given region. A method is developed for dealing with this problem and is compared with possible alternatives. The method can be used on a digital computer, and is incorporated in a program for Mercury.
Summary¶
Rosenbrock introduces a derivative-free direct-search method that adapts its coordinate directions while seeking a function maximum or minimum under constraints. The method became a classic baseline for nonlinear optimization because it is simple to implement, useful on early computers, and motivated the later Rosenbrock test function as a hard case for optimization algorithms.
Links¶
Tags¶
Similar Papers¶
1
70%
2
64%
3
63%
4
63%
5
60%