1 / 7

Evolving Worms

Evolving Worms. Concept. Implementation of classical game ”worms” Competing The one which doesn’t collide wins. Implementation Details. Static population size 500 Tournament selection (4) Byte arrays with max 255 instructions Register language

zed
Télécharger la présentation

Evolving Worms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Evolving Worms worms

  2. Concept • Implementation of classical game ”worms” • Competing • The one which doesn’t collide wins worms

  3. Implementation Details • Static population size 500 • Tournament selection (4) • Byte arrays with max 255 instructions • Register language • 2 inputs, 2 constants, 3 registers, 1 output (LEFT, RIGHT, FORWARD) • Homologous crossover 90% • Two-points-string crossover 10% • Mutation 20% worms

  4. Instruction Set • + • - • / • * • IFGT • STATE • SET worms

  5. Observations • Initally defensive tactics • Evolving to more aggressive • Eventually (800 000+ tournaments) pacifist behavior • Always turning in the same directions (after initial ambidexterity) • ”Radar” gets longer reach • Inner track wins --> U-trap --> U-trap evasion worms

  6. Future Work • Introducing mines/traps • Promote turning in both directions (two different populations, punishment of stereotype behaviour...) • Introducing apples • Growth determined by genome worms

  7. References • http://kolonisera.rymden.nu/worms worms

More Related