improving and scaling evolutionary approaches to the mastermind problem
TRANSCRIPT
- 1. J. J. Merelo , Carlos Cotta, Antonio Mora U. Granada & Mlaga (Spain) Improving and Scaling Evolutionary Approaches to the MasterMind Problem
- 2. Game of MasterMind
- 3. 7 reasons why you should care
- Donald Knuth
- 4. NP-Complete
- 5. Differential cryptanalisis/ATM cracking
- 6. Circuit/program test
- 7. Genetic profiling
- 8. Optimal solution not known
- 9. Interesting search problem
- 10. Let's play, then
- 11. Consistent combinations
- 12. Nave Algorithm
- Repeat
- Find a consistent combination and play it.
- Repeat
- 13. Looking for consistent solutions
- Optimization algorithm based on distance to consistency (for all combinations played)
- 14. Not all consistent combinations are born the same
- There's at least one better than the others (the solution).
- 15. Some will reduce the remaining search space more.
- 16. But scoring them is an open issue.
- 17. What we did before
- Play using evolutionary and co-evolutionary algorithms, fitness uses a sub-set of consistent combination
- 18. What we do now Introduceendgamesand evaluate several problem sizes
- 19. How do we use endgames? By changing our strategy after certain answers from codemaker
- 20. Endgame: All Colors Use onlypermutationsof combination
- 21. Endgame:Nullcombination Excludethose colorsfrom all combinations and reduce population accordingly.
- 22. Results
- 23. Endgamesimprove algorithmicperformance and game-playingquality
- 24. Open source your science!