Re: SQP Fix

[sidster]

Jordi Gutiérrez Hermoso-2 wrote
> On Tue, 2013-10-08 at 04:30 -0700, sidster wrote:
>> Are there any plans to fix the " ## FIXME -- check QP solution and
>> attempt
>> to recover if it has failed. ""
> [snip]
>> I checked the logs, but didn't really find any fix related to it.
> 
> It doesn't look like anyone is working on that. Are you willing to fix
> it yourself?
> 
> - Jordi G. H.

Jordi,
I am not really a maths guy, but I do have basic understanding of matrices
and how optimization generally occurs. If someone can give me detailed
instructions for implementing the recovery logic, then I can surely try, as
I have ample time. There is bug(bug#32717) filed for this problem by
Gabriele Pannocchia, in the submission this gentleman writes "I briefly
skimmed the relevant chapter of Nocedal and Wright optimization book, which
maybe was used as a reference for writing the code, and the solution does
not seem trivial and may involve using slack variables and large linear
penalties to ensure that the qp sub-problems are always feasible at each
iteration. " Do you think Nocedal and Wright optimization book is the place
to start? I am not even sure, if I will understand it. Can you provide me
with some pointers? Any help will be appreciated.

Best regards,
Siddharth 



--
View this message in context: 
http://octave.1599824.n4.nabble.com/SQP-Fix-tp4658006p4658012.html
Sent from the Octave - Maintainers mailing list archive at Nabble.com.