Sparsity exploitation in optimization

3 views (last 30 days)
Does the fmincon algorithm exploit sparsity of hessians and jacobians while solving optimization problem ?

Accepted Answer

SandeepKumar R
SandeepKumar R on 6 Mar 2019
A follow up question. Do they automatically detect sparsiy and switch to Large scale mode (sparsity exploited) or do we specifically have to supply some information to use this funtionality ?
  1 Comment
Alan Weiss
Alan Weiss on 6 Mar 2019
Generally, the large-scale algorithms do the conversion to sparse linear algebra internally, and the medium-scale algorthms do the conversion to full linear algebra internally. I tried to state this in the link I gave ("Furthermore, you do not need to specify any sparse matrices to use a large-scale algorithm.").
Alan Weiss
MATLAB mathematical toolbox documentation

Sign in to comment.

More Answers (1)

Alan Weiss
Alan Weiss on 6 Mar 2019
There are several fmincon algorithms, and they each exploit sparsity differently. The term "large-scale" means that an algorithm exploits sparsity, as explained in Large-Scale vs. Medium-Scale Algorithms. A brief outline of the fmincon algorithms is given here, including which algorithms are large-scale.
Alan Weiss
MATLAB mathematical toolbox documentation

Products


Release

R2018a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!