Brief Description: Simulated and studied the behavior of players with a common goal and competing interests, in a multi-agent setting with a dynamically changing payoff structure. Each agent (player) has competing interests, and so there are multiple player types with an associated rationale for each player type. Further, the setting follows a dynamically changing payoff structure in a repeated game with the players learning from their previous experiences, and so on. The evolution of their knowledge and their decision making strategy was studied using graph based visualizations.
To get a better context, please follow the link HerdResearch_WriteUp.pdf for the images of the graph visualizations, and for a much clear explanation.