Home首页 · Glossary术语

Glossary术语

Each card takes one example from the exercises and follows it from the table of numbers to the mistake a quick reading would make. The link at the bottom opens that page.每张卡片从练习里取一个例子，从那张数字表讲到读得太快会犯的错。底下的链接打开那一页。

Causal因果

Confounding混杂

A labour analyst has 300 workers and three columns: subsidy used or not, pay a year later, and household savings before the program. Workers with more savings both take the subsidy and earn more later. The analyst draws one line through pay and subsidy use, and the line slopes up. That slope is mixing the subsidy with the savings. The mix is confounding. If the subsidy itself adds nothing, the line can still slope up.一位劳工分析员有 300 名工人，三列数字：用没用补贴、一年后的月薪、参加前的家庭积蓄。积蓄更多的工人既去领补贴，后来月薪也更高。分析员在月薪和使用之间画一条线，线向上。这条线把补贴和积蓄搅在一起。这种搅在一起就是混杂。补贴本身没加一分钱，线仍可能向上。

Confounding lab混杂练习
Matching匹配

A department head has 80 students and three columns: joined the statistics workshop or not, a prior-math score, and the later methods grade. Students with higher prior math joined more often, so the grade gap can be that earlier score. Matching finds, for a joiner, a non-joiner with a close prior-math score, and compares those two grades. A student with no such partner is left out. The answer is then only about the pairs that could be made.一位系主任有 80 名学生，三列数字：参加统计工作坊与否、先前的数学分数、后来的方法课成绩。先前数学更高的人更常参加，所以成绩差距可能是那个旧分数。匹配为一名参加者找一名没参加、但先前数学接近的人，比较这两个人的成绩。找不到这样的人，就不进入比较。答案于是只关于能够配成的那些对。

Matching lab匹配练习
Caliper卡尺

The department head has to say how close two prior-math scores must be before the students count as a pair. That widest allowed gap is the caliper. At 2 points, many joiners find a non-joiner. At 0.3 points, most joiners find nobody, and they leave the comparison. The remaining pairs can look well matched. They are also a smaller group than the 80 students the department started with.系主任必须说，两个人的先前数学差多少以内，才算一对。这个允许的最大差距就是卡尺。差 2 分时，很多参加者找得到没参加的人。差 0.3 分时，大多数参加者找不到人，于是离开比较。剩下的配对可以看起来很接近。他们也比系里一开始的 80 人更少。

Matching lab匹配练习
Propensity倾向得分

The department head measured only prior math, and students with higher prior math were more likely to join the workshop. A propensity is one number that says how likely joining was, given the columns on the sheet. Here that number lines the 80 students up in the same order as prior math itself, because prior math is the only column. Matching on the propensity and matching on prior math then pick the same pairs.系主任只量了先前的数学，先前数学更高的学生更可能参加工作坊。倾向得分是一个数字，按表上的列，说出参加有多可能。这里这个数字把 80 名学生排成的顺序，和按先前数学排是一样的，因为先前数学是唯一的一列。按倾向得分匹配，和按先前数学匹配，会选出同样的配对。

Matching lab匹配练习

Networks网络

Stochastic block model随机块模型

A student-union survey lists 20 Arts students and 20 Engineering students, then asks each pair whether they talked last month. A stochastic block model draws those yes-or-no answers from two chances: one chance inside a faculty, another chance across faculties. The faculty names are on the sheet before any answer is drawn. The blank between the circles and the squares is where the dots were placed so the lines can be read. It is not a finding that the faculties split.学生会的调查列出文学院 20 人、工程学院 20 人，然后问每一对上个月有没有说过话。随机块模型用两个机会来画出这些是否答案：学院内部一个机会，跨学院另一个机会。学院的名字在任何答案画出之前就在表上。圆和方之间的空白，是为了把线看清楚而把点放开的位置。它不是“两个学院已经分开”这个发现。

Block-model lab块模型练习
Bridge桥

Among the 40 students, a bridge is one recorded talk between an Arts student and an Engineering student. News that an elective still has seats can pass only along recorded talks. If Engineering has many talks inside the faculty and none to Arts, the news can reach all 20 Engineering students and stop. The empty gap drawn between the two groups is not what stops it. The missing talk is.在这 40 名学生里，桥是一次记下的、文学院学生和工程学院学生之间的谈话。“选修课还有名额”只能沿记下的谈话传。如果工程学院内部谈话很多、通向文学院的一次都没有，消息可以传到全部 20 名工程学院学生，然后停下。画在两组之间的空隙不是挡住它的东西。缺了的那一次谈话才是。

How news spreads消息怎么传开
Diffusion扩散

One Engineering student says an elective still has seats. Each day that sentence can cross a recorded talk, with a chance you set, and it does not expire. Diffusion is that day-by-day passing. If the only talks are inside Engineering, the news fills those 20 students and Arts stays unmarked. A full Engineering group is not evidence that Arts is about to hear. Arts hears only after a talk crosses.工程学院的一名学生说，选修课还有名额。每一天，这句话都可以按你设定的机会跨过一次记下的谈话，而且不会过期。扩散就是这种一天一天往下传。如果谈话只在工程学院内部，消息会传满这 20 人，文学院仍然没有标记。工程学院人人都听到了，并不能说明文学院马上也会听到。文学院要等一次谈话跨过去，才会听到。

How news spreads消息怎么传开

Machine learning机器学习

Overfitting过拟合

A community-centre officer fits a curve to how willing residents were to come to activities, in 24 districts this year, using the share of households in public rental housing. The curve can bend until it touches nearly every district. Next year’s districts, drawn the same way, can sit far from that curve. The curve remembered this year’s wobbles. Overfitting is that too-close fit. A smaller error on these 24 districts is not a reason to put next year’s posters where the curve points.社区中心的同事用公屋住户占比，去拟合今年 24 个区里居民有多愿意来参加活动。曲线可以一直弯，直到几乎碰到每一个区。按同样方式再抽的明年的区，可以离这条曲线很远。曲线记住的是今年的晃动。过拟合就是这种贴得太紧。这 24 个区上的误差更小，不是把明年的活动海报贴到曲线所指之处的理由。

Overfitting lab过拟合练习
Training and validation训练与验证

Training is the 24 districts the curve was drawn to touch. Validation is a fresh batch of districts, looked at while the officer is still choosing how much the curve may bend. The curve can sit close to the 24 and far from the fresh batch. Because that batch was visible while the bend was chosen, it is still part of the choice. A last check would be districts nobody looked at until the curve was already fixed.训练，是曲线被画去贴住的那 24 个区。验证，是新的一批区，同事还在选曲线可以弯多少时就拿来看。曲线可以贴近这 24 个区，却远离新的一批。因为挑选弯曲时新的一批是看得见的，它仍是挑选的一部分。最后的核对，是曲线已经定了以后才去看的那些区。

Overfitting lab过拟合练习
Mean squared error均方误差

For each of the 24 districts the curve predicts how willing people are to come, and the prediction misses by some amount. Mean squared error averages the squared misses, so a miss of 4 counts more than four misses of 1. The number on the page is that average. A smaller number means the curve sat closer to this batch. It does not say the curve is the right one for next year’s posters.曲线为 24 个区里的每一个预测居民有多愿意来，预测都会偏一点。均方误差把这些偏差先平方再平均，所以偏了 4 的一次，比偏了 1 的四次更重。页面上的数字就是这个平均。数字更小，只表示曲线更贴近这一批。它不说明这条曲线该拿去贴明年的海报。

Overfitting lab过拟合练习
Ridge岭回归

The officer keeps a curve that may bend 10 times on the 24 districts, so every bend is still available. Ridge shortens every slope toward zero and almost never deletes one. The curve then moves less when another 24 districts are drawn. The cost is that a real bend — willingness falling only where the public-rental share is highest — can be shortened away too.同事在这 24 个区上保留一条可以弯 10 次的曲线，所以每一个弯都还在。岭回归把每个斜率往零缩短，几乎不删掉任何一个。再抽 24 个区时，曲线就少动一些。代价是真正的弯，比如愿意来的程度只在公屋占比最高的区下降，也可能被一起缩短掉。

Regularization lab正则化练习
Lasso套索

On the same curve, lasso can set a slope to exactly zero, so its bar disappears from the chart. The one-page note then names fewer bends. A missing bar means this pull dropped that bend on these 24 districts. Raise the pull, or draw next year’s districts, and the bar can return. A blank bar is not a finding that the bend never marked where people were unwilling to come.在同一条曲线上，套索可以把一个斜率正好设成零，于是它的条子从图上消失。一页说明就少写一个弯。缺掉的一条，是这个拉力在这 24 个区上丢掉了那个弯。把拉力调高，或换明年的区，这条可以回来。空白的一条，不是“这个弯从来没标出过不愿意来的区”。

Regularization lab, lasso正则化练习，套索
Penalty λ惩罚 λ

λ is how hard the penalty pulls the slopes of the 24-district curve toward zero. At 0, every slope can chase the districts. At a high value, the curve is nearly flat and ignores the public-rental share. The same λ is not the same pull if the analyst rescales that share from a fraction to a percent. The number belongs to this scaling of these columns.λ 是惩罚把这 24 个区的曲线的斜率往零拉得多用力。为 0 时，每个斜率都可以去追这些区。数值很高时，曲线几乎变平，公屋占比被忽略。如果分析员把这个占比从小数改成百分数，同一个 λ 就不是同样的拉力。这个数字属于这些列的这种缩放。

Regularization lab正则化练习
Pooled slope汇总斜率

A writing workshop has 12 students in each of four faculties, with workshop hours and a course grade. Engineering’s line can be steep and Arts’ line flat. A pooled slope ignores the faculty name and draws one line through all 48. That line often lies between the four and matches none of them. Reporting it as “the effect of the workshop” describes a student who is not in any of the four faculties.写作工作坊在四个学院各有 12 名学生，表上是工作坊时数和课程成绩。工程学院的线可以很陡，文学院的线很平。汇总斜率不管学院的名字，给全部 48 人画一条线。这条线常常落在四条线中间，哪一条都不像。把它说成“工作坊的作用”，说的是一个并不在这四个学院里的学生。

Group-slopes lab分组斜率练习
Prior先验

A hall tutor already has last year’s 24 feedback forms on whether to keep the overnight study room open, and about 30 percent said yes. That pile, and how heavily it should count, is the prior: what was on the desk before this week’s 12 chats. If the 24 forms count as a heavy pile, 8 new yeses out of 12 barely move the share. If the pile counts as light, those 12 chats move it at once.书院导师已经有去年 24 份反馈表，问通宵自修室要不要继续开，大约三成写了要。这一叠，以及它该算多重，就是先验：本周 12 次聊天到来之前，桌上已经有的看法。如果这 24 份算得很重，12 人里新的 8 个“要”几乎推不动这个比例。如果这一叠算得很轻，这 12 次聊天立刻就把它推走。

Updating lab更新练习
Posterior后验

The 12 chats arrive: 8 want the room kept open, 4 do not. The posterior is last year’s 24 forms after those 12 have been allowed to move the share. With a heavy pile of old forms, Monday’s number stays near 30 percent. With a light pile, or with 60 chats instead of 12, Monday’s number sits near 8 out of 12. A different stack of last year’s forms would have landed somewhere else. The solid curve is that updated share.12 次聊天到了：8 人希望自修室继续开，4 人不希望。后验是去年的 24 份表在这 12 次被允许推动比例之后的位置。旧表很重时，周一写下的数字留在三成附近。旧表很轻，或问的是 60 人而不是 12 人时，周一的数字靠近 12 人里的 8 人。换一叠去年的表，它会落在别处。实线就是这个更新后的比例。

Updating lab更新练习

Text as data文字当数据

Bag of words词袋

A classmate has two headlines for a tutorial note: “The new training program raised wages for the workers in the plant,” and the same line with “lowered” instead of “raised.” A bag of words counts each word and throws away the order. The two headlines then share almost every count, and the one verb that flips the claim is a small part of the score. A high score can be written up as “the headlines agree.” The counts never saw which verb came first.一个同学要为导修笔记读两则标题：“The new training program raised wages for the workers in the plant”，以及把 raised 换成 lowered 的同一句。词袋只数每个词出现几次，并把顺序丢掉。两则标题于是几乎共享每一个计数，把意思说反的那个动词只占分数的一小部分。高分可以被写成“两则标题说的是一回事”。计数从没看见哪个动词在前。

Bag-of-words lab词袋练习
Document–term matrix文档—词矩阵

Put the two headlines in a table. Each row is one headline. Each column is one word, such as training, wages, raised, or lowered. The cell is how many times that headline used that word. The two rows look alike because they share training, wages, and workers. The table does not record that one headline says pay went up and the other says pay went down. A document–term matrix is that table.把两则标题放进一张表。每一行是一则标题。每一列是一个词，比如 training、wages、raised 或 lowered。格子是这则标题用了这个词几次。两行看起来像，因为它们都有 training、wages 和 workers。表没有记下：一则说工资涨了，另一则说工资降了。文档—词矩阵就是这张表。

Bag-of-words lab词袋练习
Cosine similarity余弦相似度

Cosine is the agreement score from the two rows of word counts. It asks whether the rows point the same way, and it ignores which headline is longer. The raised/lowered pair scores high because both rows are full of the, training, program, wages, and workers. The verb that flips the claim barely moves the number. A high cosine means those counts look alike. It is not a reading of whether pay rose.余弦是从两行词数得到的一致分数。它只问两行是不是朝同一个方向，并且不管哪则标题更长。raised 和 lowered 那一对分数高，因为两行都堆满了 the、training、program、wages 和 workers。把意思说反的那个动词几乎挪不动这个数。余弦高，是这些计数看起来像。不是读出了工资到底涨没涨。

Bag-of-words lab词袋练习
Topic mixture主题混合

A classmate has nine short lines to sort before a tutorial, and has to choose how many folders to use. A topic mixture would let one line be part wage story and part curve story at the same time. This page does not do that. It gives each line one color. With one folder, wage words, talk words, and curve words share a list. With three folders, the colors can match the three lists the lines were written as. The folder count was chosen. It was not found in the lines.一个同学在导修前要给九行短句分文件夹，得先决定用几个。主题混合会让一行同时算工资的一部分，又算曲线的一部分。这一页不这么做。它给每一行一种颜色。一个文件夹时，工资的词、谈话的词和曲线的词共用一份词表。三个文件夹时，颜色可以和这九行被写成的三份清单对上。文件夹的个数是你选的。它不是从这些句子里找出来的。

Topic-count lab主题个数练习
Embedding嵌入

An embedding gives each word a place, so headlines that do not share spelling can still be scored. On this page the places were typed: raised sits opposite lowered, and pay sits near earnings. A headline’s mark is the average of its words’ places. The raised/lowered pair then sits far apart on the map even though the word counts were almost the same. A trained embedding would learn the places from many headlines. These places were typed, so the map already separates the verb that flips the claim.嵌入给每个词一个位置，于是拼写不同的标题也可以打分。这一页的位置是手写的：raised 在 lowered 对面，pay 靠近 earnings。一则标题的点，是它那些词的位置的平均。raised 和 lowered 那一对于是在图上离得很远，尽管词数几乎一样。训练出来的嵌入会从很多标题里学习这些位置。这些位置是手写的，所以地图已经把那个把意思说反的动词分开了。

Word-map lab词图练习

Glossary术语

Causal因果

Confounding混杂

Matching匹配

Caliper卡尺

Propensity倾向得分

Networks网络

Stochastic block model随机块模型

Bridge桥

Diffusion扩散

Machine learning机器学习

Overfitting过拟合

Training and validation训练与验证

Mean squared error均方误差

Ridge岭回归

Lasso套索

Penalty λ惩罚 λ

Pooled slope汇总斜率

Prior先验

Posterior后验

Text as data文字当数据

Bag of words词袋

Document–term matrix文档—词矩阵

Cosine similarity余弦相似度

Topic mixture主题混合

Embedding嵌入