eipd - simulate the ecological iterated Prisoner's Dilemma

eipd-helporeipd[-stepsinteger][-roundsinteger][-seedinteger][-CCdouble][-CDdouble][-DCdouble][-DDdouble][-Iallcdouble][-Itftdouble][-Iranddouble][-Ipavdouble][-Iallddouble][-rcpdouble][-noisedouble]

The ecological iterated Prisoner's Dilemma is simulated over time according to the specified parameters. At every time step the population of each strategy is calculated as a function of the expected scores earned against all strategies weighted by the populations of the opponents. Possible strategies include 'Always Cooperate,' 'Always Defect,'

-stepsintegerNumber of steps to simulate.-roundsintegerNumber of rounds per step.-seedintegerRandom seed for initial state.-CCdoubleReward Payoff.-CDdoubleSucker Payoff.-DCdoubleTemptation Payoff.-DDdoublePunish Payoff.-IallcdoubleInitial population of All-C.-ItftdoubleInitial population of TFT.-IranddoubleInitial population of Random.-IpavdoubleInitial population of Pavlov.-IallddoubleInitial population of All-D.-rcpdoubleProbability of C for Random strategy.-noisedoubleProbability of noise.

The payoff matrix for the Prisoner's Dilemma game is usu- ally expressed as: Player B's Move +-----------+-----------+ Player A's Move | cooperate | defect | +-----------+-----------+-----------+ | cooperate | CC, CC | CD, DC | +-----------+-----------+-----------+ | defect | DC, CD | DD, DD | +-----------+-----------+-----------+ where the table entries are (A's payoff, B's payoff) and CC, CD, DC, and DD are the reward, sucker, temptation, and punish payoffs, respectively. For each of these four out- comes you will probably want the payoffs to reflect the relationships: (DC > CC > DD > CD) and ((CD + DC) / 2 < CC).

random noise (via the -noise option) manifests itself as a cell making a randomly selected move in a single round. In this case, both the cell whose action was altered as well as that cell's opponents "remember" what the random move was on the next round. During each time step, every strategy plays against every other strategy as well as against itself. The initial population levels for all strategies will be normalized, so the scaling of the option values is irrele- vant.

No sanity checks are performed to make sure that any of the options make sense.

Copyright (c) 1997, Gary William Flake. Permission granted for any use according to the standard GNU ``copyleft'' agreement provided that the author's com- ments are neither modified nor removed. No warranty is given or implied.

Man(1) output converted with man2html