Roman Txopitea, IbaiSantana Hermida, RobertoMendiburu Alberro, AlexLozano Alonso, Jose Antonio2024-11-272024-11-27production.37422https://dx.doi.org/10.26876/ikergazte.i.95https://gordailua.ueu.eus/handle/123456789/2214Optimizazio Bayesiarra Prozesu Gaussiarren bitartez egiten denean, kernel batzuk beste batzuk bainohobeto egokitzen dira helburu-funtziora. Lan honetan, kernel hauek dinamikoki aldatzeko aukera aztertudugu, hobekuntza-probabilitatean oinarriturik.Kernelen hautaketa aurrera eramateko bost irizpideaurkeztu eta helburu-funtzio ezagunen bidez ebaluatu ditugu.Lortutako emaitzen arabera, irizpidehauek algoritmoaren errendimendua hobetzen dute kernel egokiena aurretiaz ezezaguna denean.In Bayesian Optimization, when using a Gaussian Process prior, some kernels adapt better than othersto the objective function.This research evaluates the possibility of dynamically changing the kernelfunction based on the probability of improvement. Five kernel selection strategies are proposed and testedin well known synthetic functions. According to our preliminary experiments, these methods can improvethe efficiency of the search when the best kernel for the problem is unknown.Optimizazio BayesiarraProzesu GaussiarrakOptimizazio OrokorraBayesian OptimizationGaussian ProcessGlobal OptimizationInformatikaintroduction