Optimizing search via diversity enhancement in evolutionary MasterMind
Download as ODP, PDF0 likes725 views
The document discusses optimizing search strategies in the game of Mastermind using evolutionary algorithms, focusing on enhancing diversity in search through new operators and selection mechanisms. The authors have fine-tuned evolutionary parameters to minimize evaluations and the number of games played, resulting in up to a 30% decrease in evaluations while maintaining game performance. The ultimate goals are to tackle larger sizes and improve solution quality with fewer turns.
Optimizing search via diversity enhancement in evolutionary MasterMind
- 1. Optimizing search via diversity enhancement in evolutionary MasterMind J. J. Merelo, A. Mora, C. Cotta, T. Runársson U. Granada & Málaga (Spain) & Iceland Http://geneura.wordpress.com http://twitter.com/geneura
- 2. Game of MasterMind Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 2
- 3. Let's play, then Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 3
- 4. Consistent combinations Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 4
- 5. Naïve Algorithm Repeat Find a consistent combination and play it. Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 5
- 6. Looking for consistent solutions Optimization algorithm based on distance to consistency (for all combinations played) D=2 Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 6
- 7. Not all consistent combinations are born the same There's at least one better than the others (the solution). Some will reduce the remaining search space more. But scoring them is an open issue. Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 7
- 8. What we did before Increase diversity in search via new operators and selection mechanisms Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 8
- 9. What we do now Fine-tune evolutionary parameters to minimize evaluations and number of games played Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 9
- 10. Increase diversity. Increase speed to afford tackling bigger sizes. Obtain better solutions Less turns Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 10
- 11. Consistent set size Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 11
- 12. Tournament size Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 12
- 13. Fine tuned! #Evaluations decreased up to 30%! (Game performance still the same) Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 13
- 14. Open source your science! Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 14
- 15. Thank you very much Questions? Fine tuning Evolutionary Mastermind - Merelo/Mora/Cotta/Runársson 15
Editor's Notes
- #4: How would you play mastermind? It's not easy to do, since possible branches are many more than for Sudoku or even chess. In fact, this is the kind of game that can be played more easily by a machine than by a person. CC picture from http://www.flickr.com/photos/unloveable/2399932549/
- #5: One of the possible ways to find solutions. Could be others, of course, but this is a good one.
- #8: Like the birds. They look the same, but one of them has a bad hair day. Or rather a bad feather day. Let's just say that what we do is, once a solution is consistent, we find a scoring based on how the set of consistent solutions is partitioned by comparing consistent solutions with each other. In other papers we tested different ways of doing it, and we're fixing it here. Ideally, anyways, the solution should have always the maximum fitness, but I'm not sure it does (it will have to be checked)
- #9: Creative commons image from Okinawa Soba at http://www.flickr.com/photos/24443965@N08/3606831198/ This was published in NICSO, Evostar, CIG, GECCO (as a póster) and eventually PPSN
- #10: CC Picture from San Diego Shooter http://www.flickr.com/photos/nathaninsandiego/3758988303/ New is always better. And better is also always better. Mostly.
- #11: Picture from Philip James Claxton at http://www.flickr.com/photos/philipclaxton/4076919342/in/photostream/
- #14: Image from John Traynor at http://www.flickr.com/photos/trainor/3028243647/in/photostream/
- #15: All source, data sets, experiment results for this paper are available from Sourceforge (in fact, they were while we were doing it). Source is also available from the CPAN Perl module server worldwide, in two separate modules: the algorithm itself as the module Algorithm::Mastermind (along with other algorithms; for instance, Knuth's algorithm), and the EA in the shape of the Evolutionary Algorithm library.