You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to yal...@googlegroups.com
Hi,
Say I have a problem with quadratic objective and LMI constraints.
How does Yalmip employ the objective -- as an LMI or an SOC (I use Mosek)? Which is generally more efficient?
Thanks,
Shahar.
Johan Löfberg
unread,
Aug 19, 2013, 12:38:50 PM8/19/13
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to yal...@googlegroups.com
Quadratic objective in a general conic model has to be placed in the constraint in mosek. Since Mosek supports SOCPs, there is no reason to lift the SOCP to an SDP. If that was the case, it would signal a poor implementation of the SOCP code compared to the SDP code. Hopefully Erling agrees...
Shahar Kovalsky
unread,
Aug 19, 2013, 4:17:57 PM8/19/13
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to yal...@googlegroups.com
So, just to be sure I understand correctly; In this case Yalmip will do the "right" thing and generate a second order cone constraint (corresponding to the objective epigraph)?
Erling D. Andersen
unread,
Aug 20, 2013, 1:36:16 AM8/20/13
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to yal...@googlegroups.com
Anything that can be formulated using SOC should definitely NOT be cast using the semidefinite cone. That will lead to major inefficiencies.
Johan Löfberg
unread,
Aug 20, 2013, 1:41:20 AM8/20/13
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to yal...@googlegroups.com
Yes.
Shahar Kovalsky
unread,
Aug 20, 2013, 3:29:00 AM8/20/13
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message