July 2016


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
William G Kennedy <[log in to unmask]>
Reply To:
MASON Multiagent Simulation Toolkit <[log in to unmask]>
Mon, 25 Jul 2016 19:25:41 +0000
text/plain (91 lines)
Congratulations, Sean!

On 7/25/16, 3:11 PM, "MASON Multiagent Simulation Toolkit on behalf of
Sean Luke" <[log in to unmask] on behalf of
[log in to unmask]> wrote:

>GMU has won a 3-year NSF CRI for ECJ, our evolutionary computation and
>metaheuristics toolkit (genetic algorithms, etc.).
>Because NSF told me in no uncertain terms that (of course) they would NOT
>award our team two parallel CRIs :-) this means that our NSF CRI proposal
>for MASON will not be awarded, at least this year.
>Still, you might be interested below.
>Begin forwarded message:
>> From: Sean Luke <[log in to unmask]>
>> Subject: ECJ toolkit wins National Science Foundation grant
>> Date: July 25, 2016 at 2:40:47 PM EDT
>> To: ECJ Evolutionary Computation Toolkit
>><[log in to unmask]>, genetic programming
>><[log in to unmask]>
>> George Mason University has won a 3-year NSF grant specifically to
>>enhance the ECJ evolutionary computation toolkit into a broad-based
>>library useful to the entire metaheuristics community.
>> We need two things:
>> 	1. We need suggestions for what we should do beyond the list below
>>(which is a summary of roughly what we proposed).  In order to convert
>>ECJ into a general-purpose toolkit that could serve as a central library
>>to the generl metaheuristics community, what is missing from our
>>proposed work below?  What would you like to see?
>> 	2. We'll be building a board of "power users" of ECJ to assess the
>>work and make recommendations here and there.  It'll be infrequent and
>>minimal work on your part, but it's important for us.  I would like you
>>to recommend (directly to me) people to be on that board.  Yes you can
>>recommend yourself.
>> Here's what we have proposed.  What do you think should also be there?
>> ----------------------------------------------------------------------
>> 1. Make ECJ More Robust
>> 	1A. Construct a test harness for ECJ
>> 	1B. Make distribution-based system tests for ECJ
>> 	1C. Make unit tests for ECJ
>> 2. Add Metaheuristics Frameworks to ECJ
>> 	2A. Build an Efficient Single-State optimization package (for
>>hill-climbing simualted annealing, tabu search, etc.)
>> 	2B. Build a Combinatorial Optimization package (for GRASP, AS, ACS,
>>and likely certain more recent ACO algorithms -- suggestions?)
>> 	2C. Abstract and extend the Model Fitting Package beyond CMA-ES
>>(probably to PBIL, UMDA, BIPOP-CMA-ES or AMaLGaM IDEA -- suggestions?)
>> 	2D. Build tools to permit hybrid architectures (ILS, memetic
>>algorithms), building off if our mete-evolution facility. Suggestions of
>>specific algorithms to include?
>> 	2E. Add NEAT
>> 3. Make ECJ Easier to Use
>> 	3A. Eclipse integration (wizards, code skeletons)
>> 	3B. Very significantly revise and update ECJ's GUI
>> 4. Make ECJ More Useful for Analysis
>> 	4A. Facilities to dump statistics directly to R.
>> 	4B. Integrate standard implementations of statistical analyses, such
>>as T-tests etc.
>> 	4C. Add significant number of vector benchmarks, from BBOB etc.
>> 	4D. Work with representation-specific communities (notably GP, ACO,
>>NEAT) to add often-used benchmark problems.
>> Sean