Particle Swarm Optimization with Variable Population Size

Lanzarini, Laura Cristina; Leza, Victoria; Giusti, Armando De

doi:10.1007/978-3-540-69731-2_43

Cited by 23 publications

(19 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One important feature of our method is that the algorithm uses random values in order to ensure the movement of the particle is not excessively deterministic. The most important characteristic of this paper's results is the accuracy improvement with respect to what is reported in Lanzarini et al (2008). This accuracy improvement is due to set of fuzzy classification rules, which stems from micro and macroeconomic variables.…”

mentioning

confidence: 64%

“…On the other hand, if too many individuals are used, convergence time will increase. This inconvenience could by bypassed using the variable population PSO (VarPSO) proposed by Lanzarini et al (2008). This method allows one to control the number of particles throughout the adaptive process, using concepts such as lifetime and neighborhood.…”

Section: Algorithm 1 Pseudocode Of Basic Pso Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Variations of Particle Swarm Optimization for Obtaining Classification Rules Applied to Credit Risk in Financial Institutions of Ecuador

Santana

Lanzarini

Bariviera

2019

Risks

Self Cite

View full text Add to dashboard Cite

Knowledge generated using data mining techniques is of great interest for organizations, as it facilitates tactical and strategic decision making, generating a competitive advantage. In the special case of credit granting organizations, it is important to clearly define rejection/approval criteria. In this direction, classification rules are an appropriate tool, provided that the rule set has low cardinality and that the antecedent of the rules has few conditions. This paper analyzes different solutions based on Particle Swarm Optimization (PSO) techniques, which are able to construct a set of classification rules with the aforementioned characteristics using information from the borrower and the macroeconomic environment at the time of granting the loan. In addition, to facilitate the understanding of the model, fuzzy logic is incorporated into the construction of the antecedent. To reduce the search time, the particle swarm is initialized by a competitive neural network. Different variants of PSO are applied to three databases of financial institutions in Ecuador. The first institution specializes in massive credit placement. The second institution specializes in consumer credit and business credit lines. Finally, the third institution is a savings and credit cooperative. According to our results, the incorporation of fuzzy logic generates rule sets with greater precision.

show abstract

mentioning

confidence: 64%

Section: Algorithm 1 Pseudocode Of Basic Pso Methodsmentioning

confidence: 99%

Variations of Particle Swarm Optimization for Obtaining Classification Rules Applied to Credit Risk in Financial Institutions of Ecuador

Santana

Lanzarini

Bariviera

2019

Risks

Self Cite

View full text Add to dashboard Cite

show abstract

“…[1,20,3,14,16,9,25,12]). An exception to this approach is the work of Auger and Hansen [2] in which the population size of a CMA-ES algorithm is doubled each time it is restarted.…”

Section: Related Work and Discussionmentioning

confidence: 99%

Incremental Social Learning in Particle Swarms

Stützle

Enden

et al. 2011

IEEE Trans. Syst., Man, Cybern. B

116

View full text Add to dashboard Cite

We have recently proposed two algorithms for continuous optimization that combine aspects of social learning and swarm intelligence. One of these algorithms, called IPSO, is a particle swarm optimizer with a growing population size in which new particles are initialized using a rule that biases their initial random position towards the best so far solution. The other algorithm, called IPSOLS, is an extension of IPSO which allows particles to improve their previous best position by local search.In this paper, we improve the efficiency of IPSOLS by using the local search procedure only when it is expected to be profitable. We assess its performance by comparing it with fixed and growing population size particle swarm optimization algorithms with and without local search as well as with a random restart local search algorithm. Moreover, we measure the contribution of the algorithms' core components to their overall performance on a family of problems with different fitness distance correlations.The results of the conducted experiments and of the analysis of the initialization rule show that an incremental growth of the population size in combination with a local search procedure produces a competitive algorithm that is capable of finding good solutions very rapidly without compromising its global search capabilities on problems with positive fitness distance correlation.

show abstract

“…A high value of ω at the beginning of the evolution allows the particles to make large movements, placing themselves in different positions of the search space. As the number of iterations advances, the value of ω reduces, which allows them to make a finer adjustment [73]. In this work, ω is defined as follows: The utilized ANN based PSO procedure is described as follows:…”

Section: Train Of Mlp and Rbf Using Pso Algorithmmentioning

confidence: 99%