[Ipopt-tickets] [Ipopt] #201: Inexact Algorithm

[Ipopt]

#201: Inexact Algorithm
----------------------------+------------------------
  Reporter:  ccoptimizer    |      Owner:  ipopt-team
      Type:  clarification  |     Status:  closed
  Priority:  normal         |  Component:  Ipopt
   Version:  3.11           |   Severity:  normal
Resolution:  fixed          |   Keywords:  Inexact
----------------------------+------------------------

Comment (by ccoptimizer):

 Thanks for the discussion.

 After researching some paper, I find the IPOPT inexact algorithm is mostly
 described here: http://www.optimization-
 online.org/DB_FILE/2011/04/2992.pdf

 It seems that they used the general-purpose preconditioner provided by
 Pardiso in IPOPT.
 That's probably the reason for your case, since a good preconditioner is
 quite problem dependent.

 I actually want to push my quadratic programming faster in IPOPT by using
 the iterative linear solver. At this point I would rather to follow other
 paper (e.g. http://www.maths.ed.ac.uk/~gondzio/reports/mtxFree.pdf)
 focusing on QP.  I will let you if I make any progress there. Thanks.

-- 
Ticket URL: <https://projects.coin-or.org/ticket/201#comment:5>
Ipopt <http://projects.coin-or.org/Ipopt>
Interior-point optimizer for nonlinear programs.