外文翻译遗传算法复习课程.docx
- 文档编号:11544372
- 上传时间:2023-03-19
- 格式:DOCX
- 页数:10
- 大小:66.83KB
外文翻译遗传算法复习课程.docx
《外文翻译遗传算法复习课程.docx》由会员分享,可在线阅读,更多相关《外文翻译遗传算法复习课程.docx(10页珍藏版)》请在冰豆网上搜索。
外文翻译遗传算法复习课程
Whatisageneticalgorithm?
●Methodsofrepresentation
●Methodsofselection
●Methodsofchange
●Otherproblem-solvingtechniques
Conciselystated,ageneticalgorithm(orGAforshort)isaprogrammingtechniquethatmimicsbiologicalevolutionasaproblem-solvingstrategy.Givenaspecificproblemtosolve,theinputtotheGAisasetofpotentialsolutionstothatproblem,encodedinsomefashion,andametriccalledafitnessfunctionthatallowseachcandidatetobequantitativelyevaluated.Thesecandidatesmaybesolutionsalreadyknowntowork,withtheaimoftheGAbeingtoimprovethem,butmoreoftentheyaregeneratedatrandom.
TheGAthenevaluateseachcandidateaccordingtothefitnessfunction.Inapoolofrandomlygeneratedcandidates,ofcourse,mostwillnotworkatall,andthesewillbedeleted.However,purelybychance,afewmayholdpromise-theymayshowactivity,evenifonlyweakandimperfectactivity,towardsolvingtheproblem.
Thesepromisingcandidatesarekeptandallowedtoreproduce.Multiplecopiesaremadeofthem,butthecopiesarenotperfect;randomchangesareintroducedduringthecopyingprocess.Thesedigitaloffspringthengoontothenextgeneration,forminganewpoolofcandidatesolutions,andaresubjectedtoasecondroundoffitnessevaluation.Thosecandidatesolutionswhichwereworsened,ormadenobetter,bythechangestotheircodeareagaindeleted;butagain,purelybychance,therandomvariationsintroducedintothepopulationmayhaveimprovedsomeindividuals,makingthemintobetter,morecompleteormoreefficientsolutionstotheproblemathand.Againthesewinningindividualsareselectedandcopiedoverintothenextgenerationwithrandomchanges,andtheprocessrepeats.Theexpectationisthattheaveragefitnessofthepopulationwillincreaseeachround,andsobyrepeatingthisprocessforhundredsorthousandsofrounds,verygoodsolutionstotheproblemcanbediscovered.
Asastonishingandcounterintuitiveasitmayseemtosome,geneticalgorithmshaveproventobeanenormouslypowerfulandsuccessfulproblem-solvingstrategy,dramaticallydemonstratingthepowerofevolutionaryprinciples.Geneticalgorithmshavebeenusedinawidevarietyoffieldstoevolvesolutionstoproblemsasdifficultasormoredifficultthanthosefacedbyhumandesigners.Moreover,thesolutionstheycomeupwithareoftenmoreefficient,moreelegant,ormorecomplexthananythingcomparableahumanengineerwouldproduce.Insomecases,geneticalgorithmshavecomeupwithsolutionsthatbaffletheprogrammerswhowrotethealgorithmsinthefirstplace!
Methodsofrepresentation
Beforeageneticalgorithmcanbeputtoworkonanyproblem,amethodisneededtoencodepotentialsolutionstothatprobleminaformthatacomputercanprocess.Onecommonapproachistoencodesolutionsasbinarystrings:
sequencesof1'sand0's,wherethedigitateachpositionrepresentsthevalueofsomeaspectofthesolution.Another,similarapproachistoencodesolutionsasarraysofintegersordecimalnumbers,witheachpositionagainrepresentingsomeparticularaspectofthesolution.Thisapproachallowsforgreaterprecisionandcomplexitythanthecomparativelyrestrictedmethodofusingbinarynumbersonlyandoften"isintuitivelyclosertotheproblemspace"(FlemingandPurshouse2002,p.1228).
Thistechniquewasused,forexample,intheworkofSteffenSchulze-Kremer,whowroteageneticalgorithmtopredictthethree-dimensionalstructureofaproteinbasedonthesequenceofaminoacidsthatgointoit(Mitchell1996,p.62).Schulze-Kremer'sGAusedreal-valuednumberstorepresenttheso-called"torsionangles"betweenthepeptidebondsthatconnectaminoacids.(Aproteinismadeupofasequenceofbasicbuildingblockscalledaminoacids,whicharejoinedtogetherlikethelinksinachain.Oncealltheaminoacidsarelinked,theproteinfoldsupintoacomplexthree-dimensionalshapebasedonwhichaminoacidsattracteachotherandwhichonesrepeleachother.Theshapeofaproteindeterminesitsfunction.)Geneticalgorithmsfortrainingneuralnetworksoftenusethismethodofencodingalso.
AthirdapproachistorepresentindividualsinaGAasstringsofletters,whereeachletteragainstandsforaspecificaspectofthesolution.OneexampleofthistechniqueisHiroakiKitano's"grammaticalencoding"approach,whereaGAwasputtothetaskofevolvingasimplesetofrulescalledacontext-freegrammarthatwasinturnusedtogenerateneuralnetworksforavarietyofproblems(Mitchell1996,p.74).
Thevirtueofallthreeofthesemethodsisthattheymakeiteasytodefineoperatorsthatcausetherandomchangesintheselectedcandidates:
flipa0toa1orviceversa,addorsubtractfromthevalueofanumberbyarandomlychosenamount,orchangeonelettertoanother.(SeethesectiononMethodsofchangeformoredetailaboutthegeneticoperators.)Anotherstrategy,developedprincipallybyJohnKozaofStanfordUniversityandcalledgeneticprogramming,representsprogramsasbranchingdatastructurescalledtrees(Kozaetal.2003,p.35).Inthisapproach,randomchangescanbebroughtaboutbychangingtheoperatororalteringthevalueatagivennodeinthetree,orreplacingonesubtreewithanother.
Figure1:
Threesimpleprogramtreesofthekindnormallyusedingeneticprogramming.Themathematicalexpressionthateachonerepresentsisgivenunderneath.
Itisimportanttonotethatevolutionaryalgorithmsdonotneedtorepresentcandidatesolutionsasdatastringsoffixedlength.Somedorepresenttheminthisway,butothersdonot;forexample,Kitano'sgrammaticalencodingdiscussedabovecanbeefficientlyscaledtocreatelargeandcomplexneuralnetworks,andKoza'sgeneticprogrammingtreescangrowarbitrarilylargeasnecessarytosolvewhateverproblemtheyareappliedto.
Methodsofselection
Therearemanydifferenttechniqueswhichageneticalgorithmcanusetoselecttheindividualstobecopiedoverintothenextgeneration,butlistedbelowaresomeofthemostcommonmethods.Someofthesemethodsaremutuallyexclusive,butotherscanbeandoftenareusedincombination.
Elitistselection:
Themostfitmembersofeachgenerationareguaranteedtobeselected.(MostGAsdonotusepureelitism,butinsteaduseamodifiedformwherethesinglebest,orafewofthebest,individualsfromeachgenerationarecopiedintothenextgenerationjustincasenothingbetterturnsup.)
Fitness-proportionateselection:
Morefitindividualsaremorelikely,butnotcertain,tobeselected.
Roulette-wheelselection:
Aformoffitness-proportionateselectioninwhichthechanceofanindividual'sbeingselectedisproportionaltotheamountbywhichitsfitnessisgreaterorlessthanitscompetitors'fitness.(Conceptually,thiscanberepresentedasagameofroulette-eachindividualgetsasliceofthewheel,butmorefitonesgetlargerslicesthanlessfitones.Thewheelisthenspun,andwhicheverindividual"owns"thesectiononwhichitlandseachtimeischosen.)
Scalingselection:
Astheaveragefitnessofthepopulationincreases,thestrengthoftheselectivepressurealsoincreasesandthefitnessfunctionbecomesmorediscriminating.Thismethodcanbehelpfulinmakingthebestselectionlateronwhenallindividualshaverelativelyhighfitnessandonlysmalldifferencesinfitnessdistinguishonefromanother.
Tournamentselection:
Subgroupsofindividualsarechosenfromthelargerpopulation,andmembersofeachsubgroupcompeteagainsteachother.Onlyoneindividualfromeachsubgroupischosentoreproduce.
Rankselection:
Eachindividualinthepopulationisassignedanumericalrankbasedonfitness,andselectionisbasedonthisrankingratherthanabsolutedifferencesinfitness.Theadvantageofthismethodisthatitcanpreventveryfitindividualsfromgainingdominanceearlyattheexpenseoflessfitones,whichwouldreducethepopulation'sgeneticdiversityandmighthinderattemptstofindanacceptablesolution.
Generationalselection:
Theoffspringoftheindividualsselectedfromeachgenerationbecometheentirenextgeneration.Noindividualsareretainedbetweengenerations.
Steady-stateselection:
Theoffspringoftheindividualsselectedfromeachgenerationgobackintothepre-existinggenepool,replacingsomeofthelessfitmembersofthepreviousgeneration.Someindividualsareretainedbetweengenerations.
Hierarchicalselection:
Individualsgothroughmultipleroundsofselectioneachgeneration.Lower-levelevaluationsarefasterandlessdiscriminating,whilethosethatsurvivetohigherlevelsareevaluatedmorerigorously.Theadvantageofthismethodisthatitreducesoverallcomputationtimebyusingfaster,lessselectiveevaluationtoweedoutthemajorityofindividualsthatshowlittleornopromise,andonlysubjectingthosewhosurvivethisinitialtesttomorerigorousandmorecomputationallyexpensivefitnessevaluation.
Methodsofchange
Onceselectionhaschosenfitindividuals,theymustberandomlyalteredinhopesofimprovingtheirfitnessforthenextgeneration.Therearetwobasicstrategiestoaccomplishthis.Thefirstandsimplestiscalledmutation.Justasmutationinlivingthingschangesonegenetoanother,somutationinageneticalgorithmcausessmallalterationsatsinglepointsinanindividual'scode.
Thesecondmethodiscalledcrossover,andentailschoosingtwoindividualstoswapsegmentsoftheircode,producingartificial"offspring"thatarecombinationsoftheirparents.Thisprocessisintendedtosimulatetheanalogousprocessofrecombinationthatoccurstochromosomesduringsexualreproduction.Commonformsofcrossoverinclu
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 外文 翻译 遗传 算法 复习 课程