[gecode-users] Arc consistency, value and variable ordering

Claudio Cesar de Sá claudio at colmeia.udesc.br
Wed May 5 12:28:40 CEST 2010


Christian:


The search or exploration is DFS or BAB,  OK, but I writing on the order of
variables and their values considered,
defined by the brancher....dynamically as Guido answered me, but following
some criteria defined
by brancher options. I am just trying to know what  are their meaning one by
one.

Thanks

cc


2010/5/5 Christian Schulte <cschulte at kth.se>

>  Sorry, your question does not make sense. Arc consistency has nothing to
> do with search order.
>
>
>
> Christian
>
>
>
> --
>
> Christian Schulte, www.ict.kth.se/~cschulte/<http://www.ict.kth.se/%7Ecschulte/>
>
>
>
> *From:* users-bounces at gecode.org [mailto:users-bounces at gecode.org] *On
> Behalf Of *Claudio Cesar de Sá
> *Sent:* Wednesday, May 05, 2010 11:05 AM
> *To:* users at gecode.org
> *Subject:* [gecode-users] Arc consistency, value and variable ordering
>
>
>
> Hi,
>
> In arc consistency algorithms, such the AC3 and its derivatives,  you need
> do two choices:
>
> 1: Which is order of these variables are to be considered?
> 2. Which is the order of their  values are assigned to the variable?
>
> These two parameters are in branch method, section 9.1., figure 9.1 ...
> I am with some doubts on their meanings, such:
> [INT\_VAR\_REGRET\_MIN\_MIN:]     With smallest min-regret.
> The min-regret of a variable is the difference between the smallest and
> second-smallest value still in the domain.
>
> For example: considering  variables X and Y ... Dom(X)={2, 5, 8, 11}
> and  Dom(Y)={1, 6, 18,  20}  .... X should be chosen
> once |2 - 5|  has a small  regret ... comparing with |1 - 6| of Y ?
>
> Other question, the parameter  INT\_VAR\_SIZE\_AFC\_MIN,  should the
> well-known as *first-fail*?
>
> Or INT\_VAR\_NONE,  first unassigned, should be the first variable found in
> the list of pair (variable, constraint) considered
> in arcs consistency algorithms?
>
> Where can I find details of these parameters accept by the Gecode? In
> int.hh are their specification
>
>
>
>
>
>
> --
> claudio
>
>
>
>    (
>    ))
>   |""|-.
>   |__|-'
>
>
> **********************************************************************
> Skype: claudio_cesar_sa
> Links:
> http://laic.u-clermont1.fr/
> http://www.colmeia.udesc.br/
> http://www2.joinville.udesc.br/~coca/<http://www2.joinville.udesc.br/%7Ecoca/>
> http://www2.joinville.udesc.br/~esp7maratona/<http://www2.joinville.udesc.br/%7Eesp7maratona/>
> Blog: http://claudiocesar.wordpress.com/
> ***********************************************************************
>



-- 
claudio



   (
   ))
  |""|-.
  |__|-'


**********************************************************************
Skype: claudio_cesar_sa
Links:
http://laic.u-clermont1.fr/
http://www.colmeia.udesc.br/
http://www2.joinville.udesc.br/~coca/<http://www2.joinville.udesc.br/%7Ecoca/>
http://www2.joinville.udesc.br/~esp7maratona/<http://www2.joinville.udesc.br/%7Eesp7maratona/>
Blog: http://claudiocesar.wordpress.com/
***********************************************************************



-- 
claudio



   (
   ))
  |""|-.
  |__|-'


**********************************************************************
Skype: claudio_cesar_sa
Links:
http://laic.u-clermont1.fr/
http://www.colmeia.udesc.br/
http://www2.joinville.udesc.br/~coca/
http://www2.joinville.udesc.br/~esp7maratona/
Blog: http://claudiocesar.wordpress.com/
***********************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gecode.org/pipermail/users/attachments/20100505/74663e07/attachment.htm>


More information about the users mailing list