On measuring and improving the quality of linkage learning in modern evolutionary algorithms applied to solve partially additively separable problems

Przewoźniczek, Michał Witold; Frej, Bartosz; Komarnicki, Marcin M.

doi:10.1145/3377930.3390242

Cited by 11 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As presented in [29], not all problem types are easy to decompose for a DSM-based linkage learning. Recently, Linkage Learning based on Local Optimisation (3LO) was proposed in [21].…”

Section: Populationmentioning

confidence: 99%

“…As presented in [29], the quality of the linkage may be the key to solve hard computational problems. To show the significance of using a diverse linkage, let us analyze an example shown in Figure 6.…”

Section: The Significance Of Linkage Quality and Diversitymentioning

confidence: 99%

“…However, based on the research results presented in [29], the number of levels in the pyramid is usually significantly higher than the number of subpopulations maintained by LT-GOMEA. Thus, when P3 and LT-GOMEA are applied to solve the problem, we may expect that P3 will use a more diverse linkage than LT-GOMEA.…”

Section: The Significance Of Linkage Quality and Diversitymentioning

confidence: 99%

“…When we consider multiple halls and each hall produces only a subgroup of paints, then the key issue is to successfully find the appropriate linkage that divides a genotype into subparts responsible for production on each of the halls. For DSM-using methods (like P3, or MO-GOMEA), such a task may be easy for some problem types [29], and hard for the other. If for these test cases, the key to finding a high-quality…”

Section: The Comparison Between Mo-p3 With Random and Smart Strategiesmentioning

confidence: 99%

“…According to [29], the methods that employ linkage learning are dependent on the quality of the linkage they use. If the quality of linkage is too low, linkagebased methods perform similarly to their competitors that do not consider genedependencies.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Multi-Objective parameter-less population pyramid for solving industrial process planning problems

Przewoźniczek

Dziurzański

Zhao

et al. 2021

Swarm and Evolutionary Computation

View full text Add to dashboard Cite

Evolutionary methods are effective tools for obtaining high-quality results when solving hard practical problems. Linkage learning may increase their effectiveness. One of the state-of-the-art methods that employ linkage learning is the Parameter-less Population Pyramid (P3). P3 is dedicated to solving singleobjective problems in discrete domains. Recent research shows that P3 is highly competitive when addressing problems with so-called overlapping blocks, which are typical for practical problems. In this paper, we consider a multi-objective industrial process planning problem that arises from practice and is NP-hard.To handle it, we propose a multi-objective version of P3. The extensive research shows that our proposition outperforms the competing methods for the considered practical problem and typical multi-objective benchmarks.

show abstract

“…As presented in [29], not all problem types are easy to decompose for a DSM-based linkage learning. Recently, Linkage Learning based on Local Optimisation (3LO) was proposed in [21].…”

Section: Populationmentioning

confidence: 99%

Section: The Significance Of Linkage Quality and Diversitymentioning

confidence: 99%

Section: The Significance Of Linkage Quality and Diversitymentioning

confidence: 99%

Section: The Comparison Between Mo-p3 With Random and Smart Strategiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations