[RBFOpt] FW: RBFOpt Digest, Vol 14, Issue 1
Wortmann, Thomas
thomas.wortmann at icd.uni-stuttgart.de
Fri Jan 22 12:52:28 EST 2021
Dear Nicolas,
If I understand you correctly, you could just feed the points you already know to RBFOpt when you start the algorithm, using "init_node_pos" and "init_node_val" in "rbfopt_algorithm".
init_node_pos : 2D numpy.ndarray[float] or None
Coordinates of points at which the function value is
known. These points will be added to the points generated by
the algorithm. If None, all the initial points will be
generated by the algorithm.
init_node_val : 1D numpy.ndarray[float] or None
Function values corresponding to the points given in
init_node_pos. Should be None if the previous argument is
None. If init_node_pos is not None but init_node_val is None,
the points in init_node_pos will be evaluated in the
initialization phase of the algorithm.
In that way, you don't have a noisy oracle, but you can use the points you already know to improve RBFOpt's internal surrogate model.
(Of course, Giacomo is the expert here, so my apologies if this is an irrelevant comment.)
Best,
Thomas
-----Original Message-----
From: RBFOpt <rbfopt-bounces at coin-or.org> On Behalf Of rbfopt-request at coin-or.org
Sent: Friday, January 22, 2021 6:00 PM
To: rbfopt at list.coin-or.org
Subject: RBFOpt Digest, Vol 14, Issue 1
*WARNING: THIS IS AN EXTERNAL EMAIL - Please do not click on any links or open any attachments from an unknown sender or an unexpected email.*
Send RBFOpt mailing list submissions to
rbfopt at list.coin-or.org
To subscribe or unsubscribe via the World Wide Web, visit
https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flist.coin-or.org%2Fmailman%2Flistinfo%2Frbfopt&data=04%7C01%7Cthomas_wortmann%40alumni.sutd.edu.sg%7C6db22b674a684bbc1d6508d8bef72dad%7C3476b776e9904f72b95062489831623d%7C0%7C0%7C637469316141273140%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=T31O2zk7F6tvS0x2%2F1aySzZs%2BWGX1qIjirsRYMsccFg%3D&reserved=0
or, via email, send a message with subject or body 'help' to
rbfopt-request at list.coin-or.org
You can reach the person managing the list at
rbfopt-owner at list.coin-or.org
When replying, please edit your Subject line so it is more specific than "Re: Contents of RBFOpt digest..."
Today's Topics:
1. RBFopt : question about noisy oracle definition (NICOLAS Frederic)
2. Re: RBFopt : question about noisy oracle definition
(Giacomo Nannicini)
----------------------------------------------------------------------
Message: 1
Date: Fri, 22 Jan 2021 12:54:47 +0000
From: NICOLAS Frederic <frederic.nicolas at ifpen.fr>
To: "rbfopt at list.coin-or.org" <rbfopt at list.coin-or.org>
Subject: [RBFOpt] RBFopt : question about noisy oracle definition
Message-ID:
<6B1B3843F218104A930BEC91C8D5551B02F0436B55 at IRNTSV-MBX1.ifp.fr>
Content-Type: text/plain; charset="iso-8859-1"
Dear RBFOpt support team,
I have been using RBFOpt for a few days and I have a question about the definition of the noisy oracle.
In my case, I have defined my simulator (which is exact and slow to execute) as the one that is called through an implementation in the obj_funct that is given as a parameter of the RbfoptUserBlackBox function.
On the other hand, I have a database of variants for which I already have the results (delivered earlier by my simulator) and I would like to build a surrogate model of these results to serve as the noisy oracle that I intend to implement as the obj_funct_noisy as the last parameter of the RbfoptUserBlackBox function.
I was wandering whether I could build this surrogate model with some RBFOpt existing function (that returns the 3 awaited values) or if I should use an external library to this purpose, like scikit-learn for instance.
Would you bring me a piece of advice on this ?
Thank you for your kind support.
Cordialement, Regards,
Fr?d?ric NICOLAS
IFP Energies nouvelles
Direction Sciences et Technologies du Num?rique / D?partement Mod?lisation Moteurs et V?hicules Digital Sciences and Technologies Division / Engine and Vehicle Modeling Department
1 et 4, avenue de Bois-Pr?au
92852 RUEIL-MALMAISON
FRANCE
tel. : +33(0)14752-6696
email : frederic.nicolas at ifpen.fr
__________________________
Avant d'imprimer, pensez ? l'environnement ! Please consider the environment before printing !
Ce message et toutes ses pi?ces jointes sont confidentiels et ?tablis ? l'intention exclusive de ses destinataires. Toute utilisation non conforme ? sa destination, toute diffusion ou toute publication, totale ou partielle, est interdite, sauf autorisation expresse. IFP Energies nouvelles d?cline toute responsabilit? au titre de ce message. This message and any attachments are confidential and intended solely for the addressees. Any unauthorised use or dissemination is prohibited. IFP Energies nouvelles should not be liable for this message.
__________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://apc01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flist.coin-or.org%2Fpipermail%2Frbfopt%2Fattachments%2F20210122%2F53e1f6eb%2Fattachment-0001.html&data=04%7C01%7Cthomas_wortmann%40alumni.sutd.edu.sg%7C6db22b674a684bbc1d6508d8bef72dad%7C3476b776e9904f72b95062489831623d%7C0%7C0%7C637469316141273140%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=gGuHe4L5zZ%2BGkGSzycauUz6B4ravnm6QOJLfeVZRV3g%3D&reserved=0>
------------------------------
Message: 2
Date: Fri, 22 Jan 2021 10:50:13 -0500
From: Giacomo Nannicini <giacomo.n at gmail.com>
To: NICOLAS Frederic <frederic.nicolas at ifpen.fr>
Cc: "rbfopt at list.coin-or.org" <rbfopt at list.coin-or.org>
Subject: Re: [RBFOpt] RBFopt : question about noisy oracle definition
Message-ID:
<CAFy=0d19-D3GK5r6NUmCa4j-M+-v4t0RSMjyCFa6P=ui7c_f7A at mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
Hi Frederic,
both are viable options. The surrogate models built by RBFOpt are not originally meant as predictors, therefore it is a bit clunky to build them -- but it can be done. They also do not give estimates of the error ranges. So I suspect it might be easier to use scikit-learn.
Regarding producing estimates of the error range (required as part of the return values of the noisy objective function), if the estimator doesn't provide a confidence interval, you can try to heuristically set it to some arbitrary error range and adjust it experimentally.
Giacomo
P.S: GitHub discussions very recently became available for the project, so you can use GitHub to ask questions now, if that's more convenient.
On Fri, Jan 22, 2021 at 10:43 AM NICOLAS Frederic <frederic.nicolas at ifpen.fr> wrote:
>
> Dear RBFOpt support team,
>
>
>
> I have been using RBFOpt for a few days and I have a question about the definition of the noisy oracle.
>
>
>
> In my case, I have defined my simulator (which is exact and slow to execute) as the one that is called through an implementation in the obj_funct that is given as a parameter of the RbfoptUserBlackBox function.
>
> On the other hand, I have a database of variants for which I already have the results (delivered earlier by my simulator) and I would like to build a surrogate model of these results to serve as the noisy oracle that I intend to implement as the obj_funct_noisy as the last parameter of the RbfoptUserBlackBox function.
>
>
>
> I was wandering whether I could build this surrogate model with some RBFOpt existing function (that returns the 3 awaited values) or if I should use an external library to this purpose, like scikit-learn for instance.
>
> Would you bring me a piece of advice on this ?
>
>
>
> Thank you for your kind support.
>
>
>
> Cordialement, Regards,
>
> Fr?d?ric NICOLAS
> IFP Energies nouvelles
> Direction Sciences et Technologies du Num?rique / D?partement
> Mod?lisation Moteurs et V?hicules
>
> Digital Sciences and Technologies Division / Engine and Vehicle
> Modeling Department
>
> 1 et 4, avenue de Bois-Pr?au
> 92852 RUEIL-MALMAISON
> FRANCE
> tel. : +33(0)14752-6696
> email : frederic.nicolas at ifpen.fr
>
>
>
> __________________________
>
> Avant d'imprimer, pensez ? l'environnement ! Please consider the environment before printing !
> Ce message et toutes ses pi?ces jointes sont confidentiels et ?tablis ? l'intention exclusive de ses destinataires. Toute utilisation non conforme ? sa destination, toute diffusion ou toute publication, totale ou partielle, est interdite, sauf autorisation expresse. IFP Energies nouvelles d?cline toute responsabilit? au titre de ce message. This message and any attachments are confidential and intended solely for the addressees. Any unauthorised use or dissemination is prohibited. IFP Energies nouvelles should not be liable for this message.
> __________________________
>
> _______________________________________________
> RBFOpt mailing list
> RBFOpt at list.coin-or.org
> https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flist
> .coin-or.org%2Fmailman%2Flistinfo%2Frbfopt&data=04%7C01%7Cthomas_w
> ortmann%40alumni.sutd.edu.sg%7C6db22b674a684bbc1d6508d8bef72dad%7C3476
> b776e9904f72b95062489831623d%7C0%7C0%7C637469316141273140%7CUnknown%7C
> TWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVC
> I6Mn0%3D%7C3000&sdata=T31O2zk7F6tvS0x2%2F1aySzZs%2BWGX1qIjirsRYMsc
> cFg%3D&reserved=0
------------------------------
Subject: Digest Footer
_______________________________________________
RBFOpt mailing list
RBFOpt at list.coin-or.org
https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flist.coin-or.org%2Fmailman%2Flistinfo%2Frbfopt&data=04%7C01%7Cthomas_wortmann%40alumni.sutd.edu.sg%7C6db22b674a684bbc1d6508d8bef72dad%7C3476b776e9904f72b95062489831623d%7C0%7C0%7C637469316141283138%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=d7%2BSi8GjBF4lUf0hR7vNhz57hPMOgTMSXETHeHlPOps%3D&reserved=0
------------------------------
End of RBFOpt Digest, Vol 14, Issue 1
*************************************
More information about the RBFOpt
mailing list