首页> 外国专利> METHOD AND APPARATUS FOR IMPROVING MODEL BASED ON PRE-TRAINED SEMANTIC MODEL

METHOD AND APPARATUS FOR IMPROVING MODEL BASED ON PRE-TRAINED SEMANTIC MODEL

机译:基于预训练的语义模型改进模型的方法和装置

摘要

The present disclosure provides a method and apparatus for improving a model based on a pre-trained semantic model, and relates to the technical fields of natural language processing and deep learning. A specific implementation solution is: based on the pre-trained semantic model, obtaining an initial improved model, where semantic result information of an input vector is determined in the initial improved model based on a hash search method; and based on a model distillation method, training the initial improved model to obtain an improved model. This solution obtains the semantic result information of the input vector by performing the hash search method on the input vector, replaces the original complex iterative calculation process of a semantic model, obtains the improved model with few model parameters and high compression ratio, and improves the processing speed of the improved model.
机译:本公开提供了一种用于改进基于预先训练的语义模型的模型的方法和装置,并涉及自然语言处理和深度学习的技术领域。 特定实现解决方案是:基于预先训练的语义模型,获得初始改进的模型,其中基于哈希搜索方法在初始改进模型中确定输入向量的语义结果信息; 基于模型蒸馏方法,训练初始改进模型以获得改进的模型。 该解决方案通过在输入向量上执行散列搜索方法来获得输入矢量的语义结果信息,取代语义模型的原始复杂迭代计算过程,获得具有少量模型参数和高压缩比的改进模型,并改善了 改进模型的处理速度。

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