学科分类
/ 1
1 个结果
  • 简介:Stochasticapproximationproblemistofindsomerootorextremumofanon-linearfunctionforwhichonlynoisymeasurementsofthefunctionareavailable.TheclassicalalgorithmforstochasticapproximationproblemistheRobbins-Monro(RM)algorithm,whichusesthenoisyevaluationofthenegativegradientdirectionastheiterativedirection.InordertoacceleratetheRMalgorithm,thispapergivesaflamealgorithmusingadaptiveiterativedirections.Ateachiteration,thenewalgorithmgoestowardseitherthenoisyevaluationofthenegativegradientdirectionorsomeotherdirectionsundersomeswitchcriterions.Twofeasiblechoicesofthecriterionsarepro-posedandtwocorrespondingflamealgorithmsareformed.Differentchoicesofthedirectionsunderthesamegivenswitchcriterionintheflamecanalsoformdifferentalgorithms.Wealsoproposedthesimultanousperturbationdifferenceformsforthetwoflamealgorithms.Thealmostsurelyconvergenceofthenewalgorithmsareallestablished.Thenumericalexperimentsshowthatthenewalgorithmsarepromising.

  • 标签: 随机逼近 共轭梯度 自适应方向 极值