Agent Learning and Strategy Invasion

Main Menu           Previous Topic                                                           Next Topic

Learning from Others

In the last topic we set up demonstrations where each agent had a fixed strategy, even when a neighbor seemed to be doing better. Such "incorrigible" agents, however, are hardly realistic. Human beings tend to learn from each other, and adopt behaviors that they see as successful. In the rest of our simulator, we implement agents that reflect the concept of learning, by switching to strategies of agents that seem to have high total payoffs.

                   Previous Slide                                                           Next Slide