Reevaluating Exponential Crossover in Differential Evolution

Tanabe, Ryoji; Fukunaga, Alex

doi:10.1007/978-3-319-10762-2_20

Cited by 23 publications

(10 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two mostly used crossover methods are binomial and exponential [4]. In this paper a binomial crossover is used because the exponential method is effective mostly in cases where there is a strong correlation between input vectors [17]. For every pair of elements x j i,k , v j i,k from the target and mutant vectors from current generation, a random integer, P k , between 0 and 1 is generated.…”

Section: Differential Evolutionmentioning

confidence: 99%

Forecasting Sovereign Credit Ratings Using Differential Evolution and Logic Aggregation in IBA Framework

Jelinek¹,

Milošević²,

Rakićević³

et al. 2021

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

The sovereign credit rating is considered as a quantified assessment of country's economic and political stability. Due to its importance and increasing amount of available information, the sovereign credit rating is considered as a hot topic in the last few years. However, the models that predict the credit ratings used by the several big credit rating agencies are unavailable, and can therefore be considered as the black boxes. In this paper, we are tackling this problem of predicting sovereign credit ratings by proposing a hybrid model based on interpolative Boolean algebra (IBA) and differential evolution (DE). Namely, we aim to obtain a logical/pseudo-logical function in IBA framework using DE metaheuristic that could underline connections of chosen macroeconomic indicators and sovereign credit ratings. Such functions are easy to interpret and able to make a subtle fuzzy gradation among countries. Country's economic indicators together with credit ratings from 2000 to 2016 are used for the model training. Acquired model is further tested on the data for 2017 and 2018.

show abstract

Section: Differential Evolutionmentioning

confidence: 99%

Forecasting Sovereign Credit Ratings Using Differential Evolution and Logic Aggregation in IBA Framework

Jelinek¹,

Milošević²,

Rakićević³

et al. 2021

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

show abstract

“…Binomial crossover (bin) and exponential crossover (exp) are most commonly used in DE. However, the performance of a DE algorithm using exponential crossover significantly depends on the ordering of variable indices [32]. Shuffled exponential crossover (sec), which applies exponential crossover after shuffling the variable indices of parent individuals, addresses this issue [7], [32].…”

Section: Differential Evolutionmentioning

confidence: 99%

“…However, the performance of a DE algorithm using exponential crossover significantly depends on the ordering of variable indices [32]. Shuffled exponential crossover (sec), which applies exponential crossover after shuffling the variable indices of parent individuals, addresses this issue [7], [32]. Algorithms S.2 -S.4 in the supplementary file show the three crossover methods, respectively.…”

Section: Differential Evolutionmentioning

confidence: 99%

Reviewing and Benchmarking Parameter Control Methods in Differential Evolution

Tanabe

Fukunaga

2020

IEEE Trans. Cybern.

Self Cite

View full text Add to dashboard Cite

Many Differential Evolution (DE) algorithms with various parameter control methods (PCMs) have been proposed. However, previous studies usually considered PCMs to be an integral component of a complex DE algorithm. Thus the characteristics and performance of each method are poorly understood. We present an in-depth review of 24 PCMs for the scale factor and crossover rate in DE and a large scale benchmarking study. We carefully extract the 24 PCMs from their original, complex algorithms and describe them according to a systematic manner. Our review facilitates the understanding of similarities and differences between existing, representative PCMs. The performance of DEs with the 24 PCMs and 16 variation operators is investigated on 24 black-box benchmark functions. Our benchmarking results reveal which methods exhibit high performance when embedded in a standardized framework under 16 different conditions, independent from their original, complex algorithms. We also investigate how much room there is for further improvement of PCMs by comparing the 24 methods with an oracle-based model, which can be considered to be a conservative lower bound on the performance of an optimal method.

show abstract

“…For current-to-best/1 and rand-to-best/1, the control parameters were set to = 0.05 and | | = [24]. We evaluated both binomial crossover and Shu ed Exponential Crossover (SEC) [15,20]. Since the BBOB benchmark set recommends the use of restart strategies, we used the restart strategy of [23].…”

Section: How Relevant Are the Target Trackingmentioning

confidence: 99%

Tpam

Tanabe

Fukunaga

2017

Proceedings of the Genetic and Evolutionary Computation Conference

Self Cite

View full text Add to dashboard Cite

While a large number of adaptive Di erential Evolution (DE) algorithms have been proposed, their Parameter Adaptation Methods (PAMs) are not well understood. We propose a Target functionbased PAM simulation (TPAM) framework for evaluating the tracking performance of PAMs. e proposed TPAM simulation framework measures the ability of PAMs to track prede ned target parameters, thus enabling quantitative analysis of the adaptive behavior of PAMs. We evaluate the tracking performance of PAMs of widely used ve adaptive DEs (jDE, EPSDE, JADE, MDE, and SHADE) on the proposed TPAM, and show that TPAM can provide important insights on PAMs, e.g., why the PAM of SHADE performs be er than that of JADE, and under what conditions the PAM of EPSDE fails at parameter adaptation.

show abstract

Reevaluating Exponential Crossover in Differential Evolution

Cited by 23 publications

References 16 publications

Forecasting Sovereign Credit Ratings Using Differential Evolution and Logic Aggregation in IBA Framework

Forecasting Sovereign Credit Ratings Using Differential Evolution and Logic Aggregation in IBA Framework

Reviewing and Benchmarking Parameter Control Methods in Differential Evolution

Tpam

Contact Info

Product

Resources

About