[rules-users] Monte Carlo and additional Optimisers forOptaPlanner

Fri Oct 25 10:58:17 EDT 2013

On 25-10-13 16:28, sirinath wrote:
>
> Option pricing and calculating the Greeks.
>
Sounds very intresting. If you publish anything about this problem, I 'd 
be intrested to read it.
Optimization in the financial space has many opportunities.
>
> Why is GP/GA poor in performance? Best is to find a way to make it 
> gain performance. Otherwise what is the use of adding just for the 
> sake of adding it.
>
In the experiments I 've seen (including that prototype), as well as the 
research competitions, GA is currently consistently inferiour to 
TS/LA/SA and other good local search approaches (not hill climbing of 
course).
I want to add it because
1) users often ask "isn't GA better?" and they want to see proof. If 
GA's are supported, the benchmarker can easily deliver that proof.
2) maybe there a trick we (and all the other researchers in those 
competitions) have overlooked
3) maybe there are use cases I haven't seen yet where GA's are indeed 
better.
But adding GA's is a low priority currently.
>
> Also it would be good to have journaling / event sourcing option for 
> the engine along side logging.
>
OptaPlanner supports logging nicely (see docs).
Drools might have journaling support IIRC.
What exactly do you want to accomplish with these features?

> On 25 Oct 2013 19:48, "ge0ffrey [via Drools]" <[hidden email] 
> </user/SendEmail.jtp?type=node&node=4026501&i=0>> wrote:
>
>     Boxbe <https://www.boxbe.com/overview> ge0ffrey [via Drools]
>     ([hidden email] </user/SendEmail.jtp?type=node&node=4026501&i=1>)
>     is not on your Guest List
>     <https://www.boxbe.com/approved-list?tc_serial=15501950199&tc_rand=525962296&utm_source=stf&utm_medium=email&utm_campaign=ANNO_MWTP&utm_content=001&token=puLsDj1EvX7MuWA3bFPhoZmCL%2BNgq1D%2BBxD0x%2Bsb04DVPJTEdc4IZL1NHBbsvrth&key=nFBR079qRgzQQobkU2l3REE3CTbw1ld%2BjQ6S%2B%2BXb3tY%3D>
>     | Approve sender
>     <https://www.boxbe.com/anno?tc_serial=15501950199&tc_rand=525962296&utm_source=stf&utm_medium=email&utm_campaign=ANNO_MWTP&utm_content=001&token=puLsDj1EvX7MuWA3bFPhoZmCL%2BNgq1D%2BBxD0x%2Bsb04DVPJTEdc4IZL1NHBbsvrth&key=nFBR079qRgzQQobkU2l3REE3CTbw1ld%2BjQ6S%2B%2BXb3tY%3D>
>     | Approve domain
>     <https://www.boxbe.com/anno?tc_serial=15501950199&tc_rand=525962296&utm_source=stf&utm_medium=email&utm_campaign=ANNO_MWTP&utm_content=001&dom&token=puLsDj1EvX7MuWA3bFPhoZmCL%2BNgq1D%2BBxD0x%2Bsb04DVPJTEdc4IZL1NHBbsvrth&key=nFBR079qRgzQQobkU2l3REE3CTbw1ld%2BjQ6S%2B%2BXb3tY%3D>
>
>
>
>     On 25-10-13 16:08, sirinath wrote:
>>
>>     Where is Drools chance hosted?
>>
>     https://github.com/droolsjbpm/drools-chance
>>
>>     Perhaps GA/GP can be a part of this?
>>
>     Yes, here's a Genetic Algoritms prototype for OptaPlanner:
>     https://github.com/elsam/optaplanner
>     Results with GA's were poor, but we're going to add it sooner or
>     later.
>     Here's the relevant issue to add it:
>     https://issues.jboss.org/browse/PLANNER-154
>
>>     Also Monte Carlo.
>>
>>     Maybe the current way to do is awkward? Is this covered in the
>>     documentation?
>>
>     simulation isn't covered in the documentation.
>
>     What kind of simulation do you want to do?
>     What's the problem definition?
>
>>     On 25 Oct 2013 17:54, "ge0ffrey [via Drools]" <[hidden email]
>>     <http://user/SendEmail.jtp?type=node&node=4026499&i=0>> wrote:
>>
>>     On 25-10-13 12:50, sirinath wrote:
>>     > Hi,
>>     >
>>     > Is it possible to provide more out of the box optimisers and
>>     solvers as part
>>     > of OptaPlanner.
>>     Yes, we're trying to add more every major release.
>>     For 6.1 we 'll probably add the CH's "cheapest insertion" and
>>     "regret
>>     insertion".
>>     If you want to add one yourself, start by looking at
>>     DefaultLocalSearchSolverPhase.java.
>>     >
>>     > Also add functionality for simulations (Monte Carlo) which
>>     would be a good
>>     > addition and fit.
>>     Simulation support would be great indeed - but the requirements
>>     aren't
>>     clearly definied yet (feedback, examples or use cases welcome).
>>
>>     It's already possible to do simulation + optimization
>>     by doing the simulation inside the score function (for example by
>>     running a monte carlo simulation)
>>     and doing the optimization with optaplanner.
>>
>>     drools-chance (very experimental currently) has a lot of
>>     constructs that
>>     can help in simulation (bayes stuff etc).
>>
>>     >
>>     > Suminda
>>     >
>>     >
>>     >
>>     > --
>>     > View this message in context:
>>     http://drools.46999.n3.nabble.com/Monte-Carlo-and-additional-Optimisers-for-OptaPlanner-tp4026495.html
>>     > Sent from the Drools: User forum mailing list archive at
>>     Nabble.com.
>>     > _______________________________________________
>>     > rules-users mailing list
>>     > [hidden email]
>>     <http://user/SendEmail.jtp?type=node&node=4026497&i=0>
>>     > https://lists.jboss.org/mailman/listinfo/rules-users
>>     >
>>
>>
>>     _______________________________________________
>>     rules-users mailing list
>>     [hidden email]
>>     <http://user/SendEmail.jtp?type=node&node=4026497&i=1>
>>     https://lists.jboss.org/mailman/listinfo/rules-users
>>
>>
>>     ------------------------------------------------------------------------
>>     If you reply to this email, your message will be added to the
>>     discussion below:
>>     http://drools.46999.n3.nabble.com/Monte-Carlo-and-additional-Optimisers-for-OptaPlanner-tp4026495p4026497.html
>>
>>     To unsubscribe from Monte Carlo and additional Optimisers for
>>     OptaPlanner, click here.
>>     NAML
>>     <http://drools.46999.n3.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>
> ------------------------------------------------------------------------
> View this message in context: Re: [rules-users] Monte Carlo and 
> additional Optimisers for OptaPlanner 
> <http://drools.46999.n3.nabble.com/Monte-Carlo-and-additional-Optimisers-for-OptaPlanner-tp4026495p4026499.html>
> Sent from the Drools: User forum mailing list archive 
> <http://drools.46999.n3.nabble.com/Drools-User-forum-f47000.html> at 
> Nabble.com.
>
>
> _______________________________________________
> rules-users mailing list
> [hidden email]  <http://user/SendEmail.jtp?type=node&node=4026500&i=0>
> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> [hidden email] <http://user/SendEmail.jtp?type=node&node=4026500&i=1>
> https://lists.jboss.org/mailman/listinfo/rules-users
>
> ------------------------------------------------------------------------
> If you reply to this email, your message will be added to the 
> discussion below:
> http://drools.46999.n3.nabble.com/Monte-Carlo-and-additional-Optimisers-for-OptaPlanner-tp4026495p4026500.html 
>
> To unsubscribe from Monte Carlo and additional Optimisers for 
> OptaPlanner, click here.
> NAML 
> <http://drools.46999.n3.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml> 
>
>
>
> ------------------------------------------------------------------------
> View this message in context: Re: [rules-users] Monte Carlo and 
> additional Optimisers forOptaPlanner 
> <http://drools.46999.n3.nabble.com/Re-rules-users-Monte-Carlo-and-additional-Optimisers-forOptaPlanner-tp4026501.html>
> Sent from the Drools: User forum mailing list archive 
> <http://drools.46999.n3.nabble.com/Drools-User-forum-f47000.html> at 
> Nabble.com.
>
>
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/rules-users/attachments/20131025/05d6f59c/attachment-0001.html