forum.alglib.net

ALGLIB forum
It is currently Sat Dec 21, 2024 5:15 pm

All times are UTC


Forum rules


1. This forum can be used for discussion of both ALGLIB-related and general numerical analysis questions
2. This forum is English-only - postings in other languages will be removed.



Post new topic Reply to topic  [ 3 posts ] 
Author Message
 Post subject: getting started with Linear (or Quadratic) Programming
PostPosted: Wed Apr 13, 2022 3:59 pm 
Offline

Joined: Wed Apr 13, 2022 3:09 pm
Posts: 2
Please point me to relevant examples and/or APIs in ALGLIB.

I want to find the vector x whose objective function
  • minimizes th L1 distance of x from a constant vector x0
  • where L1 distance = sum of absolute values = sum abs((x(i)-x0(i))

and subject to this set of linear constraints:
  • x(i+1) - x(i) >= constant(i) for i = 1..n-1 (i.e. x values in ascending order, with specified minimum spacing)
  • x(1) >= min_constant
  • x(n) <= max_constant

If I can get this working, then I want to try alternative distance metrics
  • L2 distance = sum of squared differences = sum (x(i)-x0(i)**2
  • L-infinity distance = max of absolute values = max abs((x(i)-x0(i))

Eventually, I want to extend the problem to two dimensions, with each x a point in the x-y plane.


Top
 Profile  
 
 Post subject: Re: getting started with Linear (or Quadratic) Programming
PostPosted: Thu Apr 14, 2022 11:54 pm 
Offline

Joined: Wed Apr 13, 2022 3:09 pm
Posts: 2
I studied the manual and think I found the right examples:
For the L1 distance objective function which is discontinuous I will try:
    Nonsmooth nonlinearly constrained optimization
    https://www.alglib.net/translator/man/manual.cpython.html#example_minns_d_nlc

For the L2 distance objective function which is quadratic I will try:
    Linearly constrained dense quadratic programming
    https://www.alglib.net/translator/man/manual.cpython.html#example_minqp_d_lc1


Top
 Profile  
 
 Post subject: Re: getting started with Linear (or Quadratic) Programming
PostPosted: Fri Jul 15, 2022 3:20 pm 
Offline

Joined: Fri Jul 15, 2022 2:24 pm
Posts: 3
I think L1 norm minimization can be cast as a linear programming problem.

Create non-negative dummy variables e_0l, e_0r, e_1l, e_1r, ...
then set equality constraints (in additional to the original constraints):
x(0) = x0(0) + e_0r
x(0) = x0(0) - e_0l
......

Then minimize e_0l + e0r + e_1l + e_1r + ...

So the LP example so be sufficient:
https://www.alglib.net/translator/man/manual.cpp.html#example_minlp_basic


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 3 posts ] 

All times are UTC


Who is online

Users browsing this forum: Google [Bot] and 16 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group