Title :
A representational sensitivity study of game theoretic simulations
Author :
Ashlock, Dan ; Leahy, Nicole
Author_Institution :
Math. Dept., Iowa State Univ., Ames, IA, USA
Abstract :
The iterated prisoner´s dilemma is a widely used computational model of cooperation and conflict. Many studies report emergent cooperation in a population of agents trained to play the prisoner´s dilemma with an evolutionary algorithm. We demonstrate levels of emergent cooperation ranging from 0% to over 90% by varying the representation of the evolving agents. The representations used in this study are finite state machines, feedforward neural networks, if-skip-action lists, parse trees storing two types of logical functions, simple look-up tables, and Markov chains. This study adds weight to the proposition that choice of representation is critical and suggests any soft computing system intended to simulate behavior must be deeply concerned with the representation issue.
Keywords :
Markov processes; feedforward neural nets; finite state machines; game theory; genetic algorithms; grammars; iterative methods; multi-agent systems; simulation; trees (mathematics); Markov chain; agent conflict; agent cooperation; agent training; behavior simulation; computational model; emergent cooperation; evolutionary algorithm; evolving agent representation; feedforward neural network; finite state machine; game theoretic simulation; if-skip-action list; iterated prisoner´s dilemma; logical function type; parse tree; representational sensitivity study; simple look-up table; soft computing system; Automata; Bioinformatics; Biological system modeling; Computational modeling; Data structures; Evolutionary computation; Feedforward neural networks; Game theory; Mathematics; Neural networks;
Conference_Titel :
Soft Computing in Industrial Applications, 2003. SMCia/03. Proceedings of the 2003 IEEE International Workshop on
Print_ISBN :
0-7803-7855-5
DOI :
10.1109/SMCIA.2003.1231346
