## Codings and operators in two genetic algorithms for the leaf-constrained minimum spanning tree problem (2004)

Venue: | International Journal of Applied Mathematics and Computer Science |

Citations: | 2 - 1 self |

### BibTeX

@ARTICLE{Julstrom04codingsand,

author = {Bryant A. Julstrom},

title = {Codings and operators in two genetic algorithms for the leaf-constrained minimum spanning tree problem},

journal = {International Journal of Applied Mathematics and Computer Science},

year = {2004},

volume = {14},

pages = {385--396}

}

### OpenURL

### Abstract

The features of an evolutionary algorithm that most determine its performance are the coding by which its chromosomes represent candidate solutions to its target problem and the operators that act on that coding. Also, when a problem involves constraints, a coding that represents only valid solutions and operators that preserve that validity represent a smaller search space and result in a more effective search. Two genetic algorithms for the leaf-constrained minimum spanning tree problem illustrate these observations. Given a connected, weighted, undirected graph G with n vertices and a bound ℓ, this problem seeks a spanning tree on G with at least ℓ leaves and minimum weight among all such trees. A greedy heuristic for the problem begins with an unconstrained minimum spanning tree on G, then economically turns interior vertices into leaves until their number reaches ℓ. One genetic algorithm encodes candidate trees with Prüfer strings decoded via the Blob Code. The second GA uses strings of length n−ℓ that specify trees ’ interior vertices. Both GAs apply operators that generate only valid chromosomes. The latter represents and searches a much smaller space. In tests on 65 instances of the problem, both Euclidean and with weights chosen randomly, the Blob-Coded GA cannot compete with the greedy heuristic, but the subsetcoded GA consistently identifies leaf-constrained spanning trees of lower weight than the greedy heuristic does, particularly on the random instances.

### Citations

532 | Shortest connection networks and some generalizations - Prim - 1957 |

440 | On the shortest spanning subtree of a graph and the traveling salesman problem - Kruskal - 1956 |

392 | A comparative analysis of selection schemes used in genetic algorithms,” in Foundations of Genetic Algorithms
- Goldberg, Deb
- 1991
(Show Context)
Citation Context ...nly 2n chromosomes. Both algorithms select chromosomes to be parents in tournaments of size two; note that these tournaments assign to chromosomes the same probabilities as linear normalization does (=-=Goldberg and Deb, 1991-=-; Julstrom, 1999). The GAs’ probability of crossover is 70% (and their probability of mutation is therefore 30%). In the Blob-Coded GA, the probability that mutation modifies any one symbol in a chrom... |

279 | OR-Library: Distributing test problems by electronic mail - Beasley - 1990 |

98 | A Theorem on Trees - Cayley |

83 | Representations for genetic and evolutionary algorithms - Rothlauf - 2006 |

70 | Neuer Beweis eines Satzes uber Permutationen - Prufer - 1918 |

45 | Representing trees in genetic algorithms
- Palmer, Kershenbaum
- 1994
(Show Context)
Citation Context ... bn−2 is a Prüfer string and T is the spanning tree. position-by-position mutation to these strings. Unfortunately, Prüfer numbers have poor properties for evolutionary search (Gottlieb et al., 2001; =-=Palmer and Kershenbaum, 1994-=-; Rothlauf and Goldberg, 1999). A small change in a Prüfer number generally changes many edges in the tree it represents, and when crossover generates a Prüfer number, the tree the offspring represent... |

35 | Greedy heuristics and an evolutionary algorithm for the bounded-diameter minimum spanning tree problem
- Raidl, Julstrom
(Show Context)
Citation Context ...uch constraints include an upper bound on trees’ degrees (Knowles and Corne, 2000; Narula and Ho, 1980; Raidl, 2000), an upper bound on trees’ diameters (Achuthan et al., 1994; Deo and Abdalla, 2000; =-=Julstrom and Raidl, 2003-=-), and the restriction of trees’ leaves to exactly two, so that valid trees are Hamiltonian paths. B.A. Julstrom A leaf of a tree is a vertex in it whose degree is one; that is, the tree connects a le... |

32 |
Algorithmic combinatorics
- Even
- 1973
(Show Context)
Citation Context ...re-parsimonious codings of valid LCSTs and operators appropriate for them. The first is based on Prüfer strings; the second specifies interior vertices. 4.1. Blob Code Cayley’s Formula (Cayley, 1889; =-=Even, 1973-=-, pp.103–4) tells us that the number of unconstrained spanning trees in a complete graph on n vertices is n n−2 . Prüfer (1918; Even, 1973, pp.104–106) presented a proof of Cayley’s Formula in the for... |

27 | Degree-constrained minimum spanning tree - Narula, Ho - 1980 |

23 | An efficient evolutionary algorithm for the degree-constrained minimum spanning tree problem
- Raidl
- 2000
(Show Context)
Citation Context ...nstraints on the spanning trees often render the search for a tree of smallest weight NP-hard. Such constraints include an upper bound on trees’ degrees (Knowles and Corne, 2000; Narula and Ho, 1980; =-=Raidl, 2000-=-), an upper bound on trees’ diameters (Achuthan et al., 1994; Deo and Abdalla, 2000; Julstrom and Raidl, 2003), and the restriction of trees’ leaves to exactly two, so that valid trees are Hamiltonian... |

21 |
An Efficient Genetic Algorithm for the p-Median Problem
- Alp, Drezner, et al.
- 2003
(Show Context)
Citation Context ...m’s elements. It has been used in evolutionary algorithms for outlier detection (Crawford et al., 1995), a problem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (=-=Alp et al., 2003-=-; Correa et al., 2001; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez, 1999; Hoelting et al., 1995; Hosage and Goodchild, 1986; Lim and Xu, 2003). In an evolutionary algorithm for the ... |

19 | Prer numbers: A poor representation of spanning trees for evolutionary search
- Gottlieb, Julstrom, et al.
- 2001
(Show Context)
Citation Context .... . , n − 1. b1b2 . . . bn−2 is a Prüfer string and T is the spanning tree. position-by-position mutation to these strings. Unfortunately, Prüfer numbers have poor properties for evolutionary search (=-=Gottlieb et al., 2001-=-; Palmer and Kershenbaum, 1994; Rothlauf and Goldberg, 1999). A small change in a Prüfer number generally changes many edges in the tree it represents, and when crossover generates a Prüfer number, th... |

14 |
Discrete Space Location-Allocation Solutions for Genetic Algorithms
- Hosage, Goodchild
- 1986
(Show Context)
Citation Context ...Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez, 1999; Hoelting et al., 1995; =-=Hosage and Goodchild, 1986-=-; Lim and Xu, 2003). In an evolutionary algorithm for the LCMST problem, a chromosome’s fitness is the total weight of the spanning tree it represents. Prim’s algorithm requires time that is O((n − ℓ)... |

14 | Edge-sets: An effective evolutionary coding of spanning trees - Raidl, Julstrom - 2003 |

13 | Otakar Boruvka on minimum spanning tree problem: translation of both the 1926 papers, comments, history - Nesetril, Milkov, et al. |

12 | How to Encode a Tree - Picciotto - 1999 |

11 | A new encoding for labeled trees employing a stack and a queue - Deo, Micikevicius |

11 | Feasible encodings for GA solutions of constrained minimal spanning tree problems - Edelson, Gargano - 2000 |

11 | A new evolutionary approach to the degree constrained minimum spanning tree problem
- Knowles, Corne
- 2000
(Show Context)
Citation Context ...in times that are polynomial in n. However, constraints on the spanning trees often render the search for a tree of smallest weight NP-hard. Such constraints include an upper bound on trees’ degrees (=-=Knowles and Corne, 2000-=-; Narula and Ho, 1980; Raidl, 2000), an upper bound on trees’ diameters (Achuthan et al., 1994; Deo and Abdalla, 2000; Julstrom and Raidl, 2003), and the restriction of trees’ leaves to exactly two, s... |

10 |
The blob code: A better string coding of spanning trees for evolutionary search
- Julstrom
- 2001
(Show Context)
Citation Context ...t vertex 0; we ignore edges’ directions to obtain undirected spanning trees. Prüfer strings decoded with the Blob Code have much better characteristics for evolutionary search than Prüfer numbers do (=-=Julstrom, 2001-=-). Further, strings decoded via the Blob Code (and the other mappings) share with Prüfer numbers a useful property. The degree of each vertex in a spanning tree is always one greater than the number o... |

8 | Computing a Diameter-constrained Minimum Spanning Tree”, PhD Dissertation, The School of Electrical Engineering and Computer - Abdalla - 2001 |

6 | Detecting multiple outliers in regression data using genetic algorithms
- Crawford, Wainwright, et al.
- 1995
(Show Context)
Citation Context ...ding is appropriate when, as here, candidate solutions are, or can be unambiguously derived from, subsets of the problem’s elements. It has been used in evolutionary algorithms for outlier detection (=-=Crawford et al., 1995-=-), a problem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez,... |

6 |
Generating interesting alternatives in GIs and SDSS using generic algorithms
- Dibble, Densham
- 1993
(Show Context)
Citation Context ...tionary algorithms for outlier detection (Crawford et al., 1995), a problem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; =-=Dibble and Densham, 1993-=-; Estivill-Castro and Torres-Velásquez, 1999; Hoelting et al., 1995; Hosage and Goodchild, 1986; Lim and Xu, 2003). In an evolutionary algorithm for the LCMST problem, a chromosome’s fitness is the to... |

6 |
Hybrid genetic algorithm for solving the p-media problem
- Estivill-Castro, Torres-Vel'azquez
- 1999
(Show Context)
Citation Context ...lier detection (Crawford et al., 1995), a problem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; Dibble and Densham, 1993; =-=Estivill-Castro and Torres-Velásquez, 1999-=-; Hoelting et al., 1995; Hosage and Goodchild, 1986; Lim and Xu, 2003). In an evolutionary algorithm for the LCMST problem, a chromosome’s fitness is the total weight of the spanning tree it represent... |

6 |
Towards solving subset selection problems with the aid of the genetic algorithm
- Lucasius, Kateman
- 1992
(Show Context)
Citation Context ...ons are, or can be unambiguously derived from, subsets of the problem’s elements. It has been used in evolutionary algorithms for outlier detection (Crawford et al., 1995), a problem in chemometrics (=-=Lucasius and Kateman, 1992-=-), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez, 1999; Hoelting et al., 1995; Hosage and Goodchild, 1986... |

4 | A heuristic for a leaf constrained minimum spanning tree problem - Deo, Micikevicius - 1999 |

3 | A study of fixed-length subset recombination - Crawford, Hoelting, et al. - 1997 |

3 |
It’s All the Same to Me: Revisiting Rank-Based Probabilities and Tournaments
- Julstrom
- 1999
(Show Context)
Citation Context ... algorithms select chromosomes to be parents in tournaments of size two; note that these tournaments assign to chromosomes the same probabilities as linear normalization does (Goldberg and Deb, 1991; =-=Julstrom, 1999-=-). The GAs’ probability of crossover is 70% (and their probability of mutation is therefore 30%). In the Blob-Coded GA, the probability that mutation modifies any one symbol in a chromosome is 2/n. Bo... |

3 | Genetic Set Recombination, Foundations of Genetic Algorithms 2 - Radcliffe - 1992 |

3 | A Study in Set Recombination - Radcliff, George - 1993 |

2 | A Genetic Algorithm for the P-median Problem
- Correa, Steiner, et al.
- 2001
(Show Context)
Citation Context ...as been used in evolutionary algorithms for outlier detection (Crawford et al., 1995), a problem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; =-=Correa et al., 2001-=-; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez, 1999; Hoelting et al., 1995; Hosage and Goodchild, 1986; Lim and Xu, 2003). In an evolutionary algorithm for the LCMST problem, a chro... |

2 |
Approximation techniques for variations of the p-median problem
- Hoelting, Schoenefeld, et al.
- 1995
(Show Context)
Citation Context ...oblem in chemometrics (Lucasius and Kateman, 1992), and particularly the p-median problem (Alp et al., 2003; Correa et al., 2001; Dibble and Densham, 1993; Estivill-Castro and Torres-Velásquez, 1999; =-=Hoelting et al., 1995-=-; Hosage and Goodchild, 1986; Lim and Xu, 2003). In an evolutionary algorithm for the LCMST problem, a chromosome’s fitness is the total weight of the spanning tree it represents. Prim’s algorithm req... |

1 | P˘rıspĕvek k ˘re˘senı otázky ekonomické stavby elektrovodnıch sıtı - O - 1926 |

1 |
ns-2 network simulator (ver.2
- unknown authors
- 2004
(Show Context)
Citation Context ...e that will increase the number of leaves, in which case the algorithm fails to find a tree that satisfies the problem’s requirements. The algorithm’s time is O(n 4 ). A more recent greedy heuristic (=-=Julstrom, 2004-=-) also begins with an unconstrained MST on G but focuses on the target graph’s vertices rather than its edges. This heuristic, which we can call ML as in “more leaves,” identifies the MST’s leaves and... |

1 | 2003): A fixed-length subset genetic algorithm for the p-median problem - Lim, Xu |