Parameter-Less Population Pyramid for Permutation-Based Problems

Woźniak, Szymon; Przewoźniczek, Michał Witold; Komarnicki, Marcin M.

doi:10.1007/978-3-030-58112-1_29

Cited by 2 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In EAs, a DSM is a square matrix that represents a degree of dependence between variables (genes) [15]. Mutual information is frequently used to estimate a DSM from the evolving population [11,28,32,35,37,38], which is defined as follows:…”

Section: Related Work 21 Dsm-based Linkage Learningmentioning

confidence: 99%

“…P4 [37] is a recently proposed version of the Parameter-less Population Pyramid (P3) [11] that was adjusted to solving permutationbased problems. Like P3, P4 employs a pyramid-like structured population rather than a classical vector (or set) structured population.…”

Section: Modern Evolutionary Permutation-based Optimizationmentioning

confidence: 99%

“…As a consequence, P4 generates many more linkage trees during a single run than LT-GOMEA, which results in a more diverse set of linkage groupings, thus increasing the exploration capacity of the Optimal Mixing operator. Both GOMEA variants, were shown competitive to other state-of-the-art methods designed to solve permutation-based problems [2,37].…”

Section: Modern Evolutionary Permutation-based Optimizationmentioning

confidence: 99%

“…We consider three different problems. The first is the Permutation Flowshop Scheduling Problem (PFSP) [31] (also considered in the papers proposing LT-GOMEA [2] and P4 [37]). Each PFSP instance is defined by 𝐽 jobs and 𝑀 machines.…”

Section: Test Problemsmentioning

confidence: 99%

“…However, the body of literature on model-based EAs for permutations is far smaller than for binary or real-valued spaces. Nevertheless, the recent advances reveal the significant potential brought by such optimizers, e.g., Generalized Mallows Estimation of Distribution Algorithm (GM-EDA) [5], Linkage Tree GOMEA (LT-GOMEA) for permutations [2], and the Parameter-less Population Pyramid for Permutation problems (P4) [37]. Therefore, in this paper, we take the next step in this promising direction.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Hybrid linkage learning for permutation optimization with Gene-pool optimal mixing evolutionary algorithms

Przewoźniczek

Komarnicki

Bosman

et al. 2021

Proceedings of the Genetic and Evolutionary Computation Conference Companion

View full text Add to dashboard Cite

Linkage learning techniques are employed to discover dependencies between problem variables. This knowledge can then be leveraged in an Evolutionary Algorithm (EA) to improve the optimization process. Of particular interest is the Gene-pool Optimal Mixing Evolutionary Algorithm (GOMEA) family, which has been shown to exploit linkage effectively. Recently, Empirical Linkage Learning (ELL) techniques were proposed for binary-encoded problems. While these techniques are computationally expensive, they have the benefit of never reporting spurious dependencies (false linkages), i.e., marking two independent variables as being dependent. However, previous research shows that despite this property, for some problems, it is more suitable to employ more commonly-used Statistical-based Linkage Learning (SLL) techniques. Therefore, we propose to use both ELL and SLL in the form of Hybrid Linkage Learning (HLL). We also propose (for the first time) a variant of ELL for permutation problems. Using a wide range of problems and different GOMEA variants, we find that also for permutation problems, in some cases, ELL is more advantageous to use while SLL is more advantageous in other cases. However, we also find that employing the proposed HLL leads to results that are better or equal than the results obtained with SLL for all the considered problems.

show abstract