|
/vol02/dtc/ | © Copyright 1995 | |||
| Volume 02 | Received: Accepted: |
---- ---- |
|||
|
The Evolution of Learning and Exploration through Improved Exploitation of Opponents
Douglas Crosher |
|
| Abstract | |
| This paper explores the hypothesis that the capacity to learn and explore can be maintained in an evolving population of interacting agents. It is demonstrated for the case of an evolving population of agents competing over a simple game. A mathematical simplification is proposed and verified against the simulated result. It is shown that a capacity to learn and explore allows individuals to correct and avoid their own mistakes, creating a large class of strategies that make no mistakes. More significantly, it is shown that, within this class, strategies which explore and learn to the greatest extent dominate due to their ability to find and exploit errors in other players. | |
| Full Text |
|
|
|
Multimedia Links (none) Reference Links (none) Citation Reference |
Get viewers for PS & PDF ![]() |