自然语言处理 (徐炀)CS3102024春  
2024春
9.0(2人评价)
  • 课程难度
    中等
  • 作业多少
    中等
  • 给分好坏
    一般
  • 收获大小
    很多
选课类别:专业任务 教学语言:英文
课程类别:专业选修课 开课单位:计算机科学与工程系
课程层次:未知 获得学分:3.0
课程主页:暂无(如果你知道,请点右上角“编辑课程信息”添加!)
课程简介(教工部数据)
This course covers all relevant knowledge in automatic speech recognition and partially introduces natural language processing. Basic concepts in machine learning, e.g. Bayesian decision theory, maximum-likelihood training, pattern classification will also be covered. The most important components in speech recognition, e.g. acoustic model, language model and pronunciation dictionary will be discussed separately. Some very popular language tasks, e.g. machine translation, information extraction, text classification will also be discussed. At the end of the course, the students should know how to design a real-life application based on the components they learned.


This course covers all relevant knowledge in automatic speech recognition and partially introduces natural language processing. Basic concepts in machine learning, e.g. Bayesian decision theory, maximum-likelihood training, pattern classification will also be covered. The most important components in speech recognition, e.g. acoustic model, language model and pronunciation dictionary will be discussed separately. Some very popular language tasks, e.g. machine translation, information extraction, text classification will also be discussed. At the end of the course, the students should know how to design a real-life application based on the components they learned.
点评写点评
排序学期
评分评分2条点评
user avatar   Cooper     2024春
  • 难度:中等
  • 作业:中等
  • 给分:一般
  • 收获:很多

涵盖了nlp的大部分基本领域,前小半个学期在讲传统nlp的技术(前transformers时代),之后就包括了transformers+LLM的内容,对于小白入门还是可以的。lab课是关于lecture的practice,难度不大,会给详细的notebook参考。lecture签到5分,lab签到5分 + practice 10分。

总共有6次Assignment(这学期最后因为内容取消了第6次)总共占55分。相关的lab和作业的参考notebook中也有代码框架。个别难度会高一些,整体难度不大。

Project占比25分,完全参考Standford CS224n: https://web.stanford.edu/class/archive/cs/cs224n/cs224n.1244/ 具体是给定代码基础上实现一个BERT + finetuned on 3 downstream multitasks。基本的实现不难,想提升performance需要自己花心思。另外还可以有customized project,需要自己选择一个topic,和老师确认一下。

没有期末考试。

一个小广告,课程内容可以参考: https://github.com/Cooper-Zhong/CS310-Natural-Language-Processing,project部分可以参考 https://github.com/Cooper-Zhong/CS310-project-minbert

user avatar   匿名用户     2024春
  • 难度:中等
  • 作业:中等
  • 给分:超好
  • 收获:很多

老师很好讲的也很清晰,project占比不大给分也很好。这个课主要还是平时作业弄好分就不会低,平时作业都按要求做也就总评扣个一两分,加上project扣个几分,而且这个课程也确实能学到一些东西


teacher avatar

徐炀

计算机科学与工程系

暂无教师主页

其他老师的「自然语言处理」课

徐炀老师的其他课