Dear colleagues,
below find the announcement for the GECCO workshop on
Real-Parameter Black-Box Optimization Benchmarking 2012.
To be notified about release of the code and documents,
please send us an email at [log in to unmask], we will include you
in our mailing list.
Don't hesitate to forward this message to possibly interested people.
Thanks,
BBOBies
----------------------------------------------------------------------
WORKSHOP ON REAL-PARAMETER BLACK-BOX OPTIMIZATION BENCHMARKING
to be held as part of the
2012 Genetic and Evolutionary Computation Conference (GECCO-2012)
July 7-11, Philadelphia, PA, USA
Organized by ACM SIGEVO
http://www.sigevo.org/gecco-2012
GECCO 2012 WORKSHOP (July 7, 2012)
Real-Parameter Black-Box Optimization Benchmarking
Submission Deadline: Wednesday, March 28th, 2012
Webpage: http://coco.gforge.inria.fr/doku.php?id=bbob-2012
Email: [log in to unmask]
Quantifying and comparing the performance of optimization algorithms
is a difficult and tedious task to achieve. Tools provided for this
workshop will ease this process by: (1) an implemented, well-motivated
single-objective benchmark function testbed, (2) a simple and sound
experimental set-up, (3) the generation of output data and (4) the
post-processing and presentation of the results in graphs and tables.
The remaining task for participants: run their favorite black-box
optimizer (old or new) by using the wrappers provided and run the
post-processing procedure (provided as well) that will generate
automatically all the material for a workshop paper (templates
are provided). A description of the algorithm and the discussion of
the results completes the task.
For the 2012 workshop edition, we provide essentially the same
test-suite as in 2010. Two testbeds are provided, (1) noise-free
functions and (2) noisy functions. The participants can freely choose
(between) both of them. This new edition also entirely bans different
parameter settings for different test functions.
This year, the post-processing tools also allow for a comparison
between more than two algorithms, for example for a well-grounded
assessment of a (new) algorithm modification. Data from BBOB 2009
contributions are also used for a comparison. The provided code also
allows for logging the restarts. The results of a portfolio approach
can now be obtained without actually running the experiment, if the
results for the constituent algorithms are available.
We especially encourage analyses that study the impact of parameter
setting changes. Besides submissions of results obtained with the
provided benchmarking setup, the workshop welcomes any submissions
concerned with benchmarking real-parameter black-box optimization
algorithms in general.
During the workshop, algorithms and results will be presented by
the participants. An overall analysis and comparison will be
accomplished by the organizers and the overall process will be
critically reviewed. A plenary discussion on future improvements will,
among others, address the question, of how the testbed should evolve.
Supporting material:
http://coco.gforge.inria.fr/
Code for the benchmark functions and for running the experiments is
provided in Matlab, C, Java, Python. An R version will be available
very soon.
Mailing list:
To subscribe to the BBOB mailing list send an email to
[log in to unmask]
Important Dates:
* 03/28/2012 paper and data submission deadline
* 04/09/2012 decision notification
* 07/07/2012 workshop
Organizers:
Anne Auger, Alexandre Chotard, Nikolaus Hansen, Verena Heidrich-Meisner,
Olaf Mersmann, Petr Posik, Mike Preuss
GECCO is sponsored by the Association for Computing Machinery Special
Interest Group on Genetic and Evolutionary Computation (SIGEVO). SIG
Services: 2 Penn Plaza, Suite 701, New York, NY, 10121, USA,
1-800-342-6626 (USA and Canada) or +212-626-0500 (Global).
|