Topological crossover for the permutation representation
Alberto Moraglio, Riccardo Poli
Abstract
Alberto Moraglio, Riccardo Poli
Abstract
Topological crossovers are a class of representation-independent operators that are well-defined once a notion of distance over the solution space is defined. In this paper we explore how the topological framework applies to the permutation representation and in particular analyze the consequences of having more than one notion of distance available. Also, we study the interactions among distances and build a rational picture in which pre-existing recombination/crossover operators for permutation fit naturally. Lastly, we also analyze the application of topological crossover to TSP.
OpenAlex reports 36 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Topological crossovers are a class of representation-independent operators that are well-defined once a notion of distance over the solution space is defined. In this paper we explore how the topological framework applies to the permutation representation and in particular analyze the consequences of having more than one notion of distance available. Also, we study the interactions among distances and build a rational picture in which pre-existing recombination/crossover operators for permutation fit naturally. Lastly, we also analyze the application of topological crossover to TSP.
Key concepts: Crossover, Representation (politics), Permutation (music), Computer science, Theoretical computer science, Topology (electrical circuits), Mathematics, Combinatorics