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]> wrote:
Boxbe ge0ffrey [via Drools] ([hidden email]) is not on your Guest List | Approve sender | Approve domain


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]> 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]
> https://lists.jboss.org/mailman/listinfo/rules-users
>


_______________________________________________
rules-users mailing list
[hidden email]
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

View this message in context: Re: [rules-users] Monte Carlo and additional Optimisers for OptaPlanner
Sent from the Drools: User forum mailing list archive at Nabble.com.


_______________________________________________
rules-users mailing list
[hidden email]
https://lists.jboss.org/mailman/listinfo/rules-users


_______________________________________________
rules-users mailing list
[hidden email]
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


View this message in context: Re: [rules-users] Monte Carlo and additional Optimisers forOptaPlanner
Sent from the Drools: User forum mailing list archive at Nabble.com.


_______________________________________________
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users