2.4 Gradient Methods



next up previous
Next: 2.4.1 Derivative Programming Up: 2 Methods for Unconstrained Previous: 2.3 Nonderivative Methods

2.4 Gradient Methods

 

Two common gradient methods are Steepest Descent (SD) [45][7] and Conjugate Gradient (CG) [36][32]. Both are fundamental techniques that are often incorporated into various iterative algorithms in many areas of scientific computing. For a theoretical treatment and notes on parallel implementations, see the linear algebra chapter gif.





verena@csep1.phy.ornl.gov