Test
Stochastic optimization algorithms
Crea una cuenta gratis para terminar este test
Guarda tu progreso y consulta tu resultado final.
¿Ya tienes cuenta? Inicia sesión
Preguntas del test (39)
1A mathematical function f(x) has always only one global minimum.
- AVerdadero
- BFalso
2Local optima occur only where f'(x) = 0.
- AVerdadero
- BFalso
3What is the feasible set (In connection with constrained optimization)? (Pick oneanswer!)
- AThe set of all points for which the function reaches its optimum.
- BThe set of all points that fulfil the constraints.
- CThe set of all points for which f(x) = 0.
4Which property applies to the Hessian matrix H (of a function f)? Pick one answer!
- AIt is symmetric
- BOnly the diagonal elements are non-zero
- CAll its elements are always non-zero
5Consider a 2x2 matrix in which all elements are equal to 1. Is this matrix positive definite?
- AVerdadero
- BFalso
6In an iterative method, once the search direction has been found, the problem of finding a suitable step length is one-dimensional.
- AVerdadero
- BFalso
7Newton-Raphson's method will always converge to a local minimum.
- AVerdadero
- BFalso
8An optimization problem is always convex if f(x) is convex.
- AVerdadero
- BFalso
9Consider the level curves of a function f = f(x1,x2) and a constraint h(x1,x2) = 0. At the localoptima of f (subject to h), the following holds:
- AThe gradients of f and h are perpendicular
- BThe gradient of h is equal to the zero vector
- CThe gradients of f and h are parallel
10Stochastic optimization methods can be applied even if the objective function f(x) is non-differentiable.
- AVerdadero
- BFalso
11 If two individuals are able to have offspring, they must be of the same species.
- AVerdadero
- BFalso
12Acquired traits (i.e. things learned) can be transmitted to the next generation.
- AVerdadero
- BFalso
13Humans have more chromosomes than any other known species.
- AVerdadero
- BFalso
14All processes in Darwinian evolution are random and without direction.
- AVerdadero
- BFalso
15In an evolutionary algorithm, the fact that chromosomes are initialized randomlyslows down the process of finding the optimum.
- AVerdadero
- BFalso
16It is important to make the right choice between binary and real-number encoding,for a givenoptimizationproblem.
- AVerdadero
- BFalso
17In tournament selection, the number of individuals participating in a tournamentis always equal to two.
- AVerdadero
- BFalso
18In some cases, after selecting two individuals, one does not carry out crossover.
- AVerdadero
- BFalso
19During mutations (in GAs) each gene is considered separately, so that the number of mutations can, in principle, be anythingfrom 0 to m(= the number of genes).
- AVerdadero
- BFalso
20The function f(j) = j(m-j), where j is the number of ones in the chromosome,is an example of a function of unitation.
- AVerdadero
- BFalso
21There are various methods to avoid premature convergence. However, another good option is simply to rerun the GA a couple of times.
- AVerdadero
- BFalso
22In LGP, regarding constants, oneis stuck with only the predefined ones,which are stored in the constant registers.
- AVerdadero
- BFalso
23In interactive evolutionary computation (IEC) one should try to present (on the screen) as many solution candidates (to the user) as possible.
- AVerdadero
- BFalso
24Biological neural networks are able to carry out complex tasks because neuronsoperate really fast.
- AVerdadero
- BFalso
25Gene regulation plays a role in long-term memory and learning.
- AVerdadero
- BFalso
26Feedforward neural networks require backpropagation for training.
- AVerdadero
- BFalso
27When fitting a model (e.g. a neural network) to a given data set, one shouldrun the training procedure for as long as possible.
- AVerdadero
- BFalso
28Ants have skilled leaders and are capable of long-distance communication
- AVerdadero
- BFalso
29What is a stigmergy
- AA mechanism for long-range communication using sound.
- BA mechanism for indirect communication bymeans of (local)modification of the environment.
- CA mechanism for long-range communication using vision.
30The movement of ants is always in the direction of the strongest pheromone scent.
- AVerdadero
- BFalso
31The change in the artificial pheromone level (in AS, applied to the TSP) on edge (i,j) dependsonlyon the (inverse of) the length of that edge.
- AVerdadero
- BFalso
32In ACO, the artificial ants always follow the edge for which the probability p(eij|S) is highest.
- AVerdadero
- BFalso
33In MMAS, one can prove that the probability of finding the best solution at least once in the first kiterations, tends to 1 as ktends to infinity. The reason that onecan prove this is the fact that (in MMAS) all edges have positivepheromone levels.
- AVerdadero
- BFalso
34In robot swarms (such as Harvard's 1000-robot swarm), only local communication(e.g. communication within the line of sight) is used.
- AVerdadero
- BFalso
35In PSO, both velocities and positions are restricted to a given range.
- AVerdadero
- BFalso
36In the simple analytical model of PSO (Appendix B.4), in which no random numbers are used (and where there is no velocity restriction), the swarm remains bounded if and only if c_1+ c_2< 4.
- AVerdadero
- BFalso
37In variable truncation PSO, all variables areintegers at all times.
- AVerdadero
- BFalso
38When comparing the performance of different settings for a stochastic optimizationalgorithm, one should make sure to use the same number of iterations (generations)for each setting.
- AVerdadero
- BFalso
39For the n-parity problem, one can separate the two classes (output = 0 and output = 1) using a single hyperplane.
- AVerdadero
- BFalso
Información
- Descripción
- Stochastic optimization algorithms Test sobre Stochastic optimization algorithms, creado por David Larsson el 04/09/2019.
- Autor
- David Larsson
- Idioma
- English
- Visitas
- 5
- Etiquetas
- Valoración
- Aún sin valoraciones Valora este test