知识卡片
静态向量合并同词的不同用法
Word2Vec、GloVe等通常为每个词项固定一个向量,将不同语境的使用混合。bank作机构与河岸若仍是同一词项,推断时得到相同向量;riverbank是另一个字符串,不能用它证明同一词项多义的完全相同表示。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 用词嵌入推理
原书出处 · 1 处
- · 查看原书依据
There are two kinds of word embeddings you may encounter in the real world: static and contextualized. Static word embeddings can be used on individual words or n-grams in isolation, and once the training is completed, the vectors remain fixed. These are the kinds of word embeddings you’ll use for analogy and other word vector reasoning problems you want to solve. You’ll train a language
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第06章《用词嵌入推理》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-6.html)
- 结论依据:摘录中关于静态向量合并同词的不同用法的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:There are two kinds of word embeddings you may encounter in the real world: static and contextualized. Static word embeddings can be used on individual words or n-grams in isolation, and once the training is completed, the vectors remain fixed. These are the kinds of word embeddings you’ll use for analogy and other word vector reasoning problems you want to solve. You’ll train a language