首页> 外国专利> method of machine learning for automatic modeling of many editions

method of machine learning for automatic modeling of many editions

机译:多版本自动建模的机器学习方法

摘要

a method and a system are revealed, a model to train a machine learning algorithm which is implemented.the technique uses latent deskriptorvektoren,a problem with many issues in a problem with spending and shows the steps to implement singlet to capture a set of training data,.process of the set of training data with a model to a set of ausgabevektoren to produceand set a set of modellparametern and komponentenwerte for at least one of the several latent latent deskriptorvektor deskriptorvektoren depending on the rate of ausgabevektoren.the set of training data has a number of eingabevektoren and several - ausgabevektoren,each of the several eingabevektor eingabevektoren is a certain latent deskriptorvektor of several latent deskriptorvektoren).each latent deskriptorvektor includes multiple scalar values on the training of the model to be initialized.
机译:揭示了一种方法和系统,一个用于训练机器学习算法的模型。该技术使用了潜在的学习技巧,该问题在支出问题中存在很多问题,并展示了实现单重态以捕获一组训练数据的步骤使用模型对一组训练数据进行处理,以生成一组ausgabevektoren,并根据ausgabevektoren的速率为多个潜在潜伏者中的至少一个潜伏者设置并设置一组模型参数和komponentenwerte。多个eingabevektoren和几个-ausgabevektoren,每个eingabevektor中的每个eingabevektoren都是几个潜伏者的特定潜伏者。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号