泛化Generalization
Generalization
解决什么问题What problem it solves
模型对没见过的数据也表现好，这才叫学会了。A model has truly learned only when it performs well on data it has never seen.
这是什么What it is
在训练数据上表现好不算数，在新数据上表现好才算。
衡量它的办法：留一部分数据不参与训练，专门用来考试。
Doing well on training data proves nothing; doing well on new data is what counts.
How it is measured: hold part of the data out of training and use it purely as an exam.
为什么需要知道Why it matters
AI 产品宣传的分数，只有来自没见过的数据才有意义。这是判断评测真假的第一关。Scores advertised by AI products only mean something if they come from unseen data. This is the first test of whether an evaluation is real.
别名Chinese aliases
举一反三
现在能做什么What you can do now
回到名词地图看第1站的其他 34 个词 →Back to the map: the other 34 terms in station 1 →