next up previous contents
Next: Golden Section Search Up: Theory: Optimization Methods Previous: Introduction

Minimization in One Dimension

The simplest minimization problems are those in which there is only a single variable. Later we will see how minimization in many dimensional space can be broken down into a series of 1-D `line' minimizations.

One major advantage of 1-D over multi-D minimization is that it is easy to define a region in which we are sure a minima must exist.

If we can find three points a, b and c with a<b<c and then there must exist at least one minimum point in the interval . The points a , b and c are said to bracket the minimum.





Bob Fisher
Fri Mar 28 14:12:50 GMT 1997