Review Questions 733(c)Roulette wheel selection process
(d)Pheromone evaporation rate
(e)Neural network
(f)Fuzzy feasible domain
(g)Membership function
(h)Multilayer feedforward network13.2 Match the following terms:
(a)Fuzzy optimization Based on shortest path
(b)Genetic algorithms Analysis equations not programmed
(c)Neural network method Linguistic data can be used
(d)Simulated annealing Based on the behavior of a flock of birds
(e)Particle swarm optimization Based on principle of survival of the fittest
(f)Ant colony optimization Based on cooling of heated solids13.3 Answer true or false:
(a)GAs can be used to solve problems with continuous design variables.
(b)GAs do not require derivatives of the objective function.
(c)Crossover involves swapping of the binary digits between two strings.
(d)Mutation operator is used to produce offsprings.
(e)No new strings are formed in the reproduction stage in GAs.
(f)Simulated annealing can be used to solve only discrete optimization problems.
(g)Particle swarm optimization is based on cognitive and social learning rates of groups
of birds.
(h)Particle swarm optimization method uses the positions and velocities of particles.
(i)Genetic algorithms basically maximize an unconstrained function.
(j)Simulated annealing basically solves an unconstrained optimization problem.
(k)GAs seek to find a better design point from a trial design point.
(l)GAs can solve a discrete optimization problem with no additional effort.
(m)SA is a type of random search technique.
(n)GAs and SA can find the global minimum with high probability.
(o)GAs are zeroth-order methods.
(p)Discrete variables need not be represented as binary strings in GAs.
(q)SA will find a local minimum if the feasible space is nonconvex.
(r)The expressions relating the input and output are to be programmed in neural-
network-based methods.
(s)Several networks architectures can be used in neural-network-based optimization.
(t)A fuzzy quantity is same as a random quantity.
(u)Ant colony optimization solves only discrete optimization problems.
(v)Fuzzy optimization involves the maximization of the intersection of the objective
function and feasible domain.13.4 Give brief answers:
(a)What is Boltzmann’s probability distribution?