Algorithmic Mechanism Design of Evolutionary Computation
Algorithm 2
Nash strategy equilibrium-based DE algorithm. fitness(): fitness function of individual ; PS: population size; Dim: dimension; : generation; maxIter: maximum generation; : index of individual; : index of dimension; gen is jump number that controls Nash equilibrium strategy investigation by several generation; in our experimental evaluation, we set gen = 1.
(1) generate an initial population.
(2) evaluate the fitness for each individual.
(3) assigning strategies for each individual.
(4) for = 1 to maxIter do
(5) for = 1 to PS do
(6) = of
(7) = of
(8) = of
(9) = rand(1, Dim)
(10) for = 1 to Dim do
(11) if rand0,1) < orthen
(12) = +
(13) =
(14) else
(15) =
(16) end if
(17) end for
(18) end for
(19) for = 1 to PS do
(20) if fitness < fitness) then
(21) replace with
(22) end if
(23) end for
(24) if mod(, gen) == 0 then
(25) calculating payoff matrix
(26) calculating Nash strategy equilibrium
(27) for = 1 to PS do
(28) if strategy of ≠ Nash strategy equilibrium then
