Lecture18

# Lecture18 - Advanced Mathematical Programming IE417 Lecture...

This preview shows pages 1–6. Sign up to view the full content.

Advanced Mathematical Programming IE417 Lecture 18 Dr. Ted Ralphs

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
IE417 Lecture 18 1 Reading for This Lecture Sections 8.8-8.9
IE417 Lecture 18 2 Methods for Large Problems Conjugate gradient methods are based on the same idea of deﬂecting the gradient to get conjugate directions. However, they use a much simpler scheme. These methods are generally not as robust and are less eﬃcient than quasi-Newton methods. However, they are much more practical for large problems. Quasi-Newton methods are generally impractical for more than 100 variables.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
IE417 Lecture 18 3 Conjugate Gradient Methods Idea : Let the next search direction depend on the last one, i.e. d j +1 = -∇ f ( y j +1 ) + α j d j As before, we require that directions produced be H-conjugate when f is quadratic. There are various choices for α j , depending on the assumptions one makes. However, all choices coincide for quadratic functions when performing exact line search.
IE417 Lecture 18 4 Fletcher-Reeves Method

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 11

Lecture18 - Advanced Mathematical Programming IE417 Lecture...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online