MathGroup Archive 2008

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: convex FindMinimum for a large vector

  • To: mathgroup at smc.vnet.net
  • Subject: [mg85809] Re: convex FindMinimum for a large vector
  • From: Frank Iannarilli <frankeye at cox.net>
  • Date: Sat, 23 Feb 2008 04:22:57 -0500 (EST)
  • References: <fpehcs$jof$1@smc.vnet.net>

Hi Art,

I presume you're using Version 6?  It sounds like you know what you're
doing, but perhaps haven't come across the various options for
FindMinimum?  Paste the following into the Help>Documentation Center
address field:

  tutorial/UnconstrainedOptimizationOverview

which should bring up a jump-table of links, including the one
pertaining to FindMinimum:
    "Methods for Local Minimization"

Perhaps tweaking the L-BFGS "step memory" or switchng to the the
ConjugateGradient method?

Good luck!


  • Prev by Date: Gudermannian Function
  • Next by Date: Re: Extracting the value from Solve
  • Previous by thread: convex FindMinimum for a large vector
  • Next by thread: Extracting the value from Solve