HyperMu’NmGA
Model was written in NetLogo 6.2.0
•
Viewed 568 times
•
Downloaded 50 times
•
Run 0 times
Do you have questions or comments about this model? Ask them here! (You'll first need to log in.)
Info tab cannot be displayed because of an encoding error
Comments and Questions
Please start the discussion about this model!
(You'll first need to log in.)
Click to Run Model
;; _______________________________________________________________________________________________________________ ;; ;; -------------------- HyperMu'NmGA --------------------------------------------------------- HyperMu'NmGA ;; HyperMu'NmGA --------------------------------------------------------- HyperMu'NmGA --------------------- ;; _______________________________________________________________________________________________________________ globals [best_result ;; the numerical value of the chromosome closer to the solution worst_result ;; the numerical value of the chromosome more distant from the solution eureka-time ;; the ticks elapsed to find the optimal solution eureka! ;; binary variable turning on when the optimal solution has just reached reduction-time;; the ticks required to get a 30% reduction of the mutator genes values average donor ;; the turtle having the best chromosome recipient ;; a random turtle different from the donor chromcopy ;; the copy of the chromosome carried by the donor turtle mutcopy ;; the copy of the mutaror gene carried by the donor turtle mate_swcopy ;; the copy of the mate_switch gene carried by the donor turtle mutator_swcopy;; the copy of the mutator_switch gene carried by the donor turtle max-mut.av ;; the highest mutators average value recorded before the problem solution min-mut.av ;; the lowest mutators average value recorded before the problem solution a-split ;; the position of one extreme in a donor's chromosome fragment b-split ;; the position of the second extreme a in a donor's chromosome fragment counter ;; a counter usefull in different blocks ] turtles-own [chromosome ;; a string of digits representing a candidate solution to the problem mutator ;; a gene able to impact positively or negatively on the genetic stability of the ;; chromosome and of itself; its strength is quantified by a whole number mutator_switch;; a gene able to determine the mutation mode; mate_switch ;; a gene able to determine the reproduction mode; ;; it harbours a binary variable: 0 for asexual reproduction, 1 for sexual reproduction fitness ;; the numerical value of the chromosomal strings (if string-value fitness-function mode) ;; or the Hamming distance complement (i.e. the proximity) to the best chromosomal ;; string (if hamming fitness-function mode) ] ;; ---------- SETUP PROCEDURES ------------------------------------------------------------------------------- ;; --------------------------------------------------------------------------------------------------------------- to setup ;; reset parameters and create a number of turtles set by the user clear-all set counter 0 reset-ticks create-turtles turtles-number [ set shape "circle 2" set size 2.3 fd random 15 genotype/fenotype-construction ] message0 set eureka-time 0 end to genotype/fenotype-construction ;; the genotype (chromosome) and the corresponding phenotype (fitness) of the turtle are initially equal to the ;; first answer that is usually given by children to the problem: "what is the greatest number of n digits?" set fitness 10 ^ (genes-number - 1) ;; the initial formal fitness value is converted into a string: it will be the chromosome structure set chromosome word fitness "" set label chromosome ;; chromosome is displayed as label ;; in the "hamming" mode, the fitness function is replaced by the complement of the Hamming distance ;; (i.e. the "Hamming proximity") to the optimal solution if fitness-function = "hamming" [compute-fitness] set mutator 1 set max-mut.av 1 set min-mut.av 1 ;; defining mate_switch gene if reproduction = "asexual" [set mate_switch 0] if reproduction = "sexual" [set mate_switch 1] if reproduction = "sex/asex" [set mate_switch random 2] ;; defining mutator_switch gene if mutate_by = "simple mutagenesis" [set mutator_switch -1] if mutate_by = "additive mutator" [set mutator_switch 0] if mutate_by = "multiplicative mutator" [set mutator_switch 1] if mutate_by = "both mutators" [set mutator_switch random 2] end ;; ---------- RUNTIME PROCEDURES ----------------------------------------------------------------------------- ;; --------------------------------------------------------------------------------------------------------------- to search ask turtles [move] ;; thew worse and the best fitness value are detected set worst_result min [fitness] of turtles set best_result max [fitness] of turtles ;; a message is sent to output when the optimal solution is reached (the solutions are differently detected ;; depending on the chosen fitness function) if eureka-time = 0 and (best_result = 10 ^ genes-number - 1 or best_result = genes-number) [ set eureka-time ticks set eureka! 1 message1 if mutate_by = "simple mutagenesis" [stop] ] ;; max-mut.av and min-mut.av are recorded before the optimal solution is reached if eureka-time = 0 [ if max-mut.av < mean [mutator] of turtles [set max-mut.av mean [mutator] of turtles] if min-mut.av > mean [mutator] of turtles [set min-mut.av mean [mutator] of turtles] ] ;; other STOP conditions if eureka-time > 0 [set reduction-time ticks - eureka-time] if eureka-time > 0 and mean [mutator] of turtles < 0.7 [message2 stop] if eureka-time = 0 and ticks >= 750000 and max [mutator] of turtles < 1 [message3 stop] if eureka-time > 0 and min [mutator] of turtles >= 1.5 and reduction-time > 500000 [message4 stop] ;; two alternative kinds of selective reproduction are admitted: sexual and asexual ;; they take place if the diversity between all chromosomes is not null clear-links if best_result != worst_result [ set donor [who] of one-of turtles with [fitness = best_result] set chromcopy [chromosome] of turtle donor set mutcopy [mutator] of turtle donor set mate_swcopy [mate_switch] of turtle donor set mutator_swcopy [mutator_switch] of turtle donor set recipient [who] of one-of turtles with [fitness != best_result] ask turtle donor [ create-link-to turtle recipient if mate_switch = 0 [cloning] if mate_switch = 1 [recombination] ] ] ;; mutations occur randomly, the frequency is related to the basic mutation-rate (the mutagenicity ;; of the environment) and the total gene number of turtles population if random-float 1 < basic_mutation-rate * turtles-number * (genes-number + 3) [ask one-of turtles [mutation] ] tick set eureka! 0 end ;; TALK TO ME! --------------------------------------------------------------------------------------------------- to message0 output-print "*** HyperMu'NmGA ***" output-print " " end to message1 output-print " OPTIMAL SOLUTION" output-print " reached on " output-print word " tick: " eureka-time end to message2 output-print " " output-print " Mutators mean went" output-print " down 30% under the" output-print " starting value after" output-type " ticks: " output-type reduction-time end to message3 output-print " " output-print " Low evolutive" output-print " potential" end to message4 output-print " " output-print " No mutator < 1.5" output-print " possible endless" output-print " hypermutation status" end ;; GOOD VIBRATIONS ----------------------------------------------------------------------------------------------- to-report patches-ahead [rad dis] ; reports to turtles a set of patches ahead report [patches in-radius rad] of patch-ahead dis end to move ifelse (any? other turtles-on patches-ahead 1 1) [ bk 1 lt random-float 360] [fd 0.001] end ;; Operator 1. FITNESS FUNCTIONS ;; ---------------------------------------------------------------------------------------------------------------- to compute-fitness set fitness 0 set counter 0 if fitness-function = "hamming" [hamming-proximity] if fitness-function = "string-value" [ set fitness read-from-string chromosome] set label chromosome end to hamming-proximity if counter = length chromosome [set counter 0 stop] if item counter chromosome = "9" [ set fitness fitness + 1 ] set counter counter + 1 hamming-proximity end ;; Operator 2. REPRODUCTION ;; ---------------------------------------------------------------------------------------------------------------- to cloning ask turtle recipient [ set chromosome chromcopy set mutator mutcopy set mutator_switch mutator_swcopy set mate_switch mate_swcopy compute-fitness ] end to recombination ;; homologous recombination occurs between the chromosome of a randomly chosen turtle (recipient) and the ;; chromosome of (one of) the most performing one (donor) that offers a code's fragment of its chromosome ;; to the first turtle; the two involved turtles will be highlighted by a link set counter 0 set a-split random genes-number set b-split random genes-number ask turtle recipient [ hybridization compute-fitness] end to hybridization ;; the two selected chromosomal strings give place to hybridization according to a mechanism ;; similar to the crossing-over following to bacterial conjugation; strings are looped, ;; as occur usually in bacterial chromosomes or plasmids ifelse a-split < b-split [set chromosome replace-item (a-split + counter) chromosome (item (a-split + counter) chromcopy) set counter (counter + 1) if counter < b-split - a-split [hybridization]] [if b-split < a-split [set chromosome replace-item ((a-split + counter) mod genes-number) chromosome (item ((a-split + counter) mod genes-number) chromcopy) set counter (counter + 1) if counter < genes-number - a-split + b-split [hybridization]] ] ;; if required, the three donor's regulative genes are transferred into the recipient turtle if reproduction = "sex/asex" [set mate_switch 1] if mutate_by != "simple mutagenesis" [ if random-float 1 < hym-ratio [set mutator mutcopy] if mutate_by = "both mutators" [if random-float 1 < 0.5 [ set mutator_switch mutator_swcopy] ] ] end ;; Operator 3: MUTAGENESIS ;; ---------------------------------------------------------------------------------------------------------------- to mutation ;; simple mutagenesis doesn't involve mutator genes ifelse mutate_by = "simple mutagenesis" [point-mutation compute-fitness if reproduction = "sex/asex" [ ;; mutation of the reproduction mode if random-float 1 < 0.5 [set mate_switch ((mate_switch + 1) mod 2)] ] ] ;; hypermutation can produce a mutation of a structural gene as well as a mutation on the mutator gene; ;; the last one determines the mutation number along the chromosome: it can be 0, 1 or more (multimutation) [if random-float 1 > hym-ratio [multimutation compute-fitness] if random-float 1 < hym-ratio [hypermutation] ;; when the population is reproductively heterogeneous, also the mate_switch gene can mutate if reproduction = "sex/asex" [ if random-float 1 * mutator * mu-expressivity > hym-ratio [ set mate_switch ((mate_switch + 1) mod 2)] ;; mutation of the reproduction mode ] ;; when the population is mutationally heterogeneous, also the mutator_switch gene can mutate if mutate_by = "both mutators" [ if random-float 1 * mutator * mu-expressivity > hym-ratio [ set mutator_switch ((mutator_switch + 1) mod 2)] ;; mutation of the mutator mode ] ] end to point-mutation ;; point-mutations hit randomly only one gene set chromosome replace-item random genes-number chromosome word random 10 "" end to multimutation ;; the multimutations imply different point-mutations in a unique mutation event ;; the number of contemporary point-mutations is determined by the variable stored in the mutator-gene ifelse mutator < genes-number * 4 [repeat mutator [point-mutation]] ;; if the mutator value is too high (more than four times the number of the chromosome genes) in order to avoid ;; overflow events, the number of point-mutations is cut to four times the number of chromosome genes [repeat genes-number * 4 [point-mutation] ] end to hypermutation ;; mutator gene can mutate itself in two alternative ways (additive or multiplicative, decided by ;; mutator_switch gene), the self-mutation effect can be modulated also by the mutator-expressivity parameter set mutator mutator + (random 3 - 1) * mutator ^ mutator_switch * mu-expressivity end
There are 7 versions of this model.
Attached files
File | Type | Description | Last updated | |
---|---|---|---|---|
HyperMu’NmGA.png | preview | HyperMu-Reflection | about 4 years ago, by Cosimo Leuci | Download |
READ_ME.txt | data | attachment | about 1 year ago, by Cosimo Leuci | Download |
Parent: Minimal Genetic Algorithm
This model does not have any descendants.