chenghuimin.Rmd

---
title: "Breiman Page 1-5"
author: "huimincheng"
date: "Wednesday, November 02, 2016"
output: html_document
---

### A Conversation with Leo Breiman

###Leo Breiman 访谈实录

##Abstract

**1**： Leo Breiman was born in New York City on January 27, 1928.
His parents and he migrated five years later to San Francisco where he
began school. During Leo’s junior high school years, his family moved
again, to Los Angeles. Leo graduated from Roosevelt High School in 1945
and entered the California Institute of Technology, from which he graduated
four years later with a major in physics. He earned his Master’s
Degree in mathematics at Columbia University in 1950 and his Ph.D. in
Mathematics at the University of California, Berkeley in 1954.

1928年1月28日，Leo Breiman诞生于纽约，在他出生5年后他们全家移民去了旧金山，在这里他开始了学业生活。在他读初中的时候，他们全家又移民到了洛杉矶。1945年，Leo从Roosevelt高中毕业后考进了加州理工学院，在加州理工学院他花了4年时间主修物理。1950年，Leo拿到了哥伦比亚大学的数学系硕士学位，1954年，他又拿到了加州大学伯克利分校的数学系博士毕业学位。


**2**：Leo has broad ranging scientific and mathematical interests, including information theory and the theory of gambling. He has been involved in applications coming from studies of automobile traffic, air quality and toxic substance recognition. He is the author of a celebrated graduate text on probability theory, is one of four authors of Classification and Regression Trees and its associated CARTR software and has also written two other books. With Jerome Friedman, Leo developed the ACE (alternating conditional expectations) algorithm by which nonlinear relationships between the dependent variable and predictor variables in regression are described. He is the originator of “bagging” and “arcing” both computer-intensive approaches to classification that are of much current interest.

Leo对科学和数学有着广泛的兴趣，包括信息论和博弈论。他曾参与研究汽车交通，空气质量和有毒物质的识别。
他写过一篇著名的有关概率论的毕业论文，是分类回归树及其配套软件CARTR的四大作者之一，另外他还写了两本其他的书。Leo和Jerome Friedman一起发展了ACE（alternating conditional expectations）算法，该算法描述了因变量和自变量之间的非线性回归关系。
他开创性地将 "bagging" 和 "arcing" 这两种需要大量计算的方法用于分类，目前很多人对此十分感兴趣。


**3**：Leo’s professional positions have included being on the faculty of the
Department of Mathematics at UCLA, an independent consultant for 13
years and Professor of Statistics and founding Director of the Statistical
Computing Facility at the University of California, Berkeley. 
In addition,he has had visiting positions at Stanford and at Yale. For his many
contributions, Leo has been honored by Fellowship in the Institute of
Mathematical Statistics and in the American Statistical Association. He
is an elected member of the American Academy of Arts and Sciences and
received the Berkeley Citation from the University of California.

Leo的职业生涯经历了UCLA数学系的任教，13年的独立咨询顾问，在加州大学伯克利分校统计系的教授和统计计算设施的创始主任。
而且，他还是斯坦福和耶鲁大学的客座教授。
由于他的诸多贡献，Leo被授予数理统计研究所（Institute of
Mathematical Statistics ）和美国统计协会（American Statistical Association）的荣誉奖学金。
同时，他还当选美国艺术与科学学院（American Academy of Arts and Sciences）的成员，并被加州大学授予 Berkeley Citation荣誉奖项。

**4**：The interests and accomplishments of Leo Breiman extend outside the
areas of professional statistician and probabilist. He was a waiter in
the Catskills, a dishwasher in the Merchant Marine, a trekker into the
heart of rainforest Africa, an active father to many children from a small
agrarian Mexican village, a member and President of the Santa Monica
School Board, the architect of his stunning home and an accomplished
sculptor. Leo and his wife, Mary Lou, reside in Berkeley. He is the father
of two daughters, Rebecca and Jessica.

Leo Breiman的兴趣和成就不只限于专业的统计学家和概率学家
他在Catskills当过服务员，在Merchant Marine当过洗碗工，还是探寻过热带雨林之心的背包客，是一群来自墨西哥农村孩子的慈爱父亲，是Santa Monica学校董事会的主席，是他美丽小屋的建筑师，同时还是一个技艺高超的雕刻家。 Leo和他的妻子Mary Lou，居住在伯克利（Berkeley），他们育有两个女儿，Rebecca和Jessica。


Richard Olshen is Professor of Biostatics, Chief, Division of Biostatistics, Associate Chair, Department of Health Research and Policy, Professor (by courtesy)of Electrical Engineering and of Statistics, Stanford University, Division of Biostatistics, Department of Health Research and
Policy, Stanford, California 94305-5405 
e-mail:olshen@stat.stanford.edu.

Richard Olshen简介：Richard Olshen是斯坦福大学生物统计教授，生物统计科（Division of Biostatistic）主任，卫生研究和政策系（Department of Health Research and Policy）的副主席，电气工程与统计系教授。

This conversation was held at Leo and Mary Lou
Breiman’s home on February 19, 1999.
该访谈于1999年2月19日在Leo和他的妻子Mary Lou Breiman家中进行。

##正文
**Olshen**：
Leo, I want to say that it’s a privilege to
be here. I was an admirer of your work and taught
from one of your books before we collaborated, in
fact, before I knew you. I understand that you grew
up in Boyle Heights in Los Angeles. What was Boyle
Heights like?
**Olshen**：
Leo,今天在这里我感到非常荣幸。我是您的一枚粉丝，在跟您合作之前（事实上，是我认识您之前）我从您的著作中学习到了很多。我了解到您是在洛杉矶的Boyle Heights长大的，Boyle Heights那个时候是什么样的？

##BOYLE HEIGHTS
**Breiman**：
 Well, Boyle Heights was, at that
point, a poor Jewish ghetto, working-class Jewish
ghetto, which was bordered on one side by the
Mexican ghetto. It was basically a first-generation
immigrant ghetto. My parents were immigrants;
everyone else’s parents were immigrants.

Boyle Heights那时候还是个贫穷的犹太人区，工人阶级的犹太人贫民区，与墨西哥贫民区相邻。基本上，它是第一代移民的贫民区。我的父母，那里其他人的父母也是移民。

And we were all going to Roosevelt High School,
which was busy turning offspring of immigrants into
doctors and lawyers and scientists. It was a wonderful
place to be. You could go into a delicatessen, get
a pickle out of a pickle barrel, walk down the street
and see salamis hanging and drying. It was something
just a little upscale from the lower East Side
of New York.

而且我们都是去罗斯福（Roosevelt）高中上学，这所高中致力把移民的后代培养成医生、律师和科学家。
这是一个很美好的地方。
你可以走进一家熟食店，从泡菜桶里拿出泡菜，然后沿着街道一直走，可以比人晾晒看到意大利腊肠（salamis）。意大利腊肠（salamis）是来自纽约下东区（纽约市曼哈顿区沿东河南端一带，犹太移民聚居地）有点高档的食品。



**Olshen: **And better weather perhaps.

**Olshen: **也许还有更好的天气

**Breiman**：Right.

**Breiman**：是的 

**Olshen: **When did your interests in mathematics
and science begin to emerge?

**Olshen: **你什么时候把对数学和科学的兴趣结合起来的呢？

**Breiman**：I got turned on by geometry in high
school because geometry was the first subject I took
that really impressed me. You know , when you hit
geometry after having done algebra or added numbers
or something, it’s so completely different and
fascinating that a bunch of us couldn’t stop working
on it.

**Breiman**：我在高中的时候对几何感兴趣，因为几何是第一门让我印象深刻的科目。
你知道吗，当你解决完代数或者加上数字之类的问题之后，挑战几何问题是一件全然不同而且令人着迷的事，以至于我们一群人在研究它的时候根本停不下来。

**Olshen: **What more can youtell me about
Roosevelt High School, the neighborhood, what got
youinterested in Caltech and actually got youto
Caltech?

**Olshen: **关于罗斯福（Roosevelt）高中所在的社区你还有什么其他的能告诉我们的吗？是什么促使你对加州理工学院(Caltech)感兴趣并真正加入加州理工学院的呢？

**Breiman**： Well, Roosevelt High School, although
it was in one of the poorest socioeconomic neighborhoods
of Los Angeles at that time, was also one
of the premier academic high schools, probably for
the same reason that, say, the Bronx High School of
Science was.

It was filled with highly motivated sons and
daughters of immigrants, all of whom knew they
were somehow going to wind up in college. And
what stimulated me to go to Caltech was, I think,
my mother’s hearing that it was the best scientific
school in the West and deciding that her son should
go to Caltech.

**Breiman**：
嗯，虽然罗斯福(Roosevelt)社区那时候是洛杉矶最穷的社会经济社区之一，但罗斯福高中是一流的学术高中之一。
这个原因也许和Bronx科学高中的原因是一样的。

这里充满了有激情的移民家庭的儿女，大家都明确地知道一定要上大学。
而激发我去加州理工学院(Caltech)的原因，是我母亲听说加州理工学院是西方最好的科学性学校，于是她决定她儿子应该去加州理工学院。

**Olshen: ** Did you have sibs?

**Olshen: **你有兄弟姐妹吗？

**Breiman**：No.

**Breiman**：没有

**Olshen: **So it really mattered what happened to
the one, the one son?

**Olshen: **所以唯一的儿子干什么真的非常重要

**Breiman**：Right.

**Breiman**：是的

##加州理工学院(Caltech) 
**Olshen: **Tell me a little bit about Caltech) as it
was when you were there.

**Olshen: **能跟我讲一下你还在加州理工学院的时候，那儿是什么样的吗？

**Breiman**：Sure. The entering class at Caltech
then was about 250 out of the many thousands who
applied. And to get in, you had to take 16 hours
of exams. They had no faith in high school grades
at all. There was a four-hour exam in physics, four
hours in math, four hours in chemistry, and four
hours in English. And we sweated it all out.

My first year in Caltech I got a scholarship, and
I did extremely well. And then my second and
third and fourth years, I got more and more fed up
because I got so tired of hearing nothing but science
and engineering that it began to turn me off.
The place was like a scientific monastery.

So as a result, my grades started going down and
going down and going down. And in my last year, I
got four D’s in my major subject, which was physics.


**Breiman**：当然。那年申请加州理工学院的几千人，但是最终录取的只有250人。为了能被录取，你必须参与16个小时的考试，因为他们完全不相信高中成绩。
你必须考数学、物理、化学和英语四科，每科考四小时，而且我们需要把它们都写出来。

在加州理工学院的第一年，我确实表现得很好，也拿到了奖学金。但是第二、三、四年，我变得越来越烦闷，我厌倦了每天听到的不是科学就是电气，这让我开始封闭起来。这个地方就像是一个科学修道院。

所以，我的成绩开始急剧下滑，在我最后一年时，我的主修科目物理得到了4个D。


**Olshen: **And this is what commended you to
the University of California at Berkeley Graduate
School?

**Olshen: **这是让你去加州大学伯克利分校研究生院的原因吗？


**Breiman**：No. What happened was my grades in
physics were terrible, but I kept being interested in
math. I did fairly well in math. I applied all over,
and the only place that accepted me was Columbia
University in New York City. My parents were quite
poor and couldn’t afford the tuition at Columbia.


But from what I had saved from working, I was
able to go to Columbia; and I got my Master’s in
mathematics in a year. My grades in math were
pretty good, but actually I had gone to Columbia
determined to become a philosophy major.


When I got there, I talked to the head of the
philosophy department, Erwin Edwin. At the time
he was quite well known. And he said, patting me
paternalistically on the knee, “Look, two of my finest
Ph.D.’s right now can’t get jobs. Why don’t yousta y
in math and take a few philosophy courses and see
how it works out.”

So I took a few philosophy courses—for instance,
a course in aesthetics, a course in Greek philosophy
and so on. I decided this is not at all what I had in
mind. So there I was back in math.

**Breiman**：不是，实际上是虽然我毕业时的物理成绩太差了，但是我还是对数学很感兴趣。我在数学方面很好，我到处申请，最终只有纽约的哥伦比亚大学接收了我。我的父母十分贫穷，而且没法承担我在哥大的学费。

但是由于之前我打工存了一些钱，所以我最终还是有能力去哥大的。然后我用1年的时间在哥大获得了数学硕士学位，但是实际上我最开始去哥大时是想主修哲学的。

当我到了这之后，我和赫赫有名的哲学系系主任Erwin Edwin交谈了一番。他语重心长地拍了拍我，说道，“你看，我两个最优秀的博士生现在都没找到工作。你为什么不待在数学系，然后上一些哲学系的课看看呢？”

然后我就上了一些哲学系的课，比如说美学和希腊哲学等等。
最终，我还是决定回归到了数学。

**Olshen: **Tell us exactly when this was.

**Olshen: **能告诉我们具体的时间吗？


**Breiman**：This was in 1950. I got my Master’s
degree after a year at Columbia, and I applied to the
Berkeley Math Department. They said, “Come.” To
support myself, I worked one summer as a waiter in
the Catskills and for another I was in the Merchant
Marine.

**Breiman**：那是1950年的时候，我从哥大拿到硕士毕业学位1年之后，我申请了伯克利的数学系，他们接受了我。为了能支付学费，我1个暑假在Catskills（卡次启尔山脉，位于纽约的南部）当服务员打暑假工，另一个暑假在美国商船学院（Merchant Marine）。


**Olshen: **The Merchant Marine sounds good.

**Olshen: **美国商船学院（Merchant Marine）听起来很不错

##GRADUATE SCHOOL AT BERKELEY

##毕业于伯克利

**Breiman**：Merchant Marine. Right. And I made
a large amount of money for a kid in his early 20’s.
I was the dishwasher on board ship. And I got off
the ship just in time to proctor an exam that all
Berkeley Math T.A.’s were supposed to be at. I took
a taxi from the dock and I rushed over to Dwinelle
where this big exam with hundreds of freshman was
being given. So—

**Breiman**：是的，在我20岁之前，我这样打工挣了很多钱。我在船上当洗碗工，一下船就得去监考，监考是所有伯克利助教必须干的事。从码头叫了一辆出租车，我迅速地奔向大一新生考试的地方Dwinelle，所以—

**Olshen: **So you interleaved your sailing career with your mathematical career?

**Olshen: **所以你的航海生涯和数学生涯交错进行？

**Breiman**：Right. Right.

**Breiman**：确实是

**Olshen: **Excellent. Tell me a little bit about the
influence of David Blackwell and others with whom
you studied at Berkeley. I’m thinking in particular
with regard to two subjects: one of them is the
Shannon–Breiman–MacMillan theorem (Breiman,
1957) and the other is your work on gambling
(Breiman, 1960).

**Olshen: **棒极了，能跟我们讲讲David Blackwell对你的影响吗？还有其他跟你一起在伯克利学习的人也可以讲讲吗？我其实是想问的是和这两个主题相关的：一个是Shannon–Breiman–MacMillan理论(Breiman,
1957)，另外一个是你博彩系统相关的工作(Breiman, 1960)。

**Breiman**：Well, I was in the Math Department
at Berkeley at that time. There was no separate
statistics department. There were a few people at
that time who Neyman managed to bring in, such
as  Michel Loève è . And I was taking mathematics
courses, and then I took the probability course from
 Michel Loève è , and I loved it. I loved probability theory.
And that was quite a class, a probability course
of 10 people or so. Manny Parzen was in that class,
Howard Tucker, too, and one or two other people
who went on to become well known in statistics.
Then I wrote a thesis. But  Michel Loève è was a perfectionist,
kept me working on the thesis, and working
on the thesis.

**Breiman**：好的，我当时是在伯克利的数学系，那会儿还没有独立的统计系。那个时候，Neyman想引入一些人，比如 Michel Loève è 。当时，我上的概率论这门课是教的，我非常喜欢这门课，我喜欢概率论。

这是一门很好的课程，一门大概只有10个人左右的课程。Manny Parzen也在这个课堂里，还有Howard Tucker，还有一两个后来在统计界声名远扬的人。然后我写了一篇论文，但是 Michel Loève è 是一个完美主义者，他让我在论文上不断下功夫。

**Olshen: **I can’t help but be reminded that Michel Loève è  was my adviser when I was a freshman at
Berkeley. And the first words he spoke to me when
I walked in his office were, “You better make all A’sor I’m going to throw youou t of this window.” That
was his speech to a frightened freshman!

**Olshen: **我情不自禁地想起来，当我大一刚入学时，我的导师是Michel Loève è ，他在办公室对我说的第一句话就是，"你最好全部课程都得A，不然我会把你从窗户扔出去。"

**Breiman**：That sounds like Michel. So I had
written this thesis, and I thought it was pretty
good. Michel kept saying, “Well, I don’t know.”
Then, Harold Cram´er came to Berkeley for a
semester and organized a seminar. And I gave my
thesis results in his seminar.

Afterward he said, “I’ll have to talk to Michel and
tell him I think this is pretty good.” But the other
important thing that happened was I got a notice
from my draft board saying, “This is it. You’ve used
up all of your deferments. We expect to see you here
in 30 days.”


And so I handed Michel a copy of the latest version
of my thesis, with the draft notice on top. And
he took one long look, and he said, “Well, we have to
get youou t of here pretty fast.” So he organized the
final defense of the thesis in about two weeks and
put a stamp of approval on it. I was thrilled. Then
I served almost two years in the Army.

**Breiman**：这听起来很像是Michel的风格，所以每次我写完论文觉得这已经很不错的时候，Michel都会说“我不知道”。然后，Harold Cram´er来到了伯克利一学期，我在他组织的研讨会上提交了我的论文结果。

他说，“我必须跟Michel好好谈谈，我觉得这篇论文已经很不错了”。
但是，这时候一件更加重要的事情发生了，我收到了一个来自征兵局的通知，通知上说“注意！你已经用尽了所有的延期时间，我们希望在30天之后在这里看到你”

所以我给Michel提交了我最新一版论文的复件，同时把那个征兵局的通知也放在上面。Michel盯着它看了很久，然后说，“好吧，我们必须让你尽快离开这里了”。大概在两周之后，他组织了论文的最后答辩，然后我的论文获得了批准，我当时很激动。离开伯克利之后，我服役了将近两年。

**Olshen: **Wait. Let’s back up. Was your thesis
related to the Shannon–Breiman–MacMillan
theorem or to gambling?

**Olshen: **等一下，让我们回过头。你的毕业论文和Shannon–Breiman–MacMillan理论有关，还是和你的博彩系统有关呢？


**Breiman**：No. What happened was that in the
Army then, you could get out two months early if
youhad a job. I wrote to Professor Neyman and he
gave me a job as acting assistant professor. I came
back and at that time I got to know David Blackwell
pretty well.

We used to sit in seminars, and he would send
me these little notes saying, “Can youprove this
and this and this?” And one of the notes he sent
me was, “Can youprove that this thing converges
almost surely to the following?” And I thought about
it for the next couple of days. And then I thought,
“Yes. I can do that. I can see how to do that.”
That was how the Shannon–Breiman–McMillan
Theorem (Breiman, 1957) came about. Dave Blackwell
was, for me, an incredibly inspirational person
to work with. To see the way his mind worked was
terrific.

**Breiman**：不是的，当时军队有个规定是，如果你能找到工作，你就可以提前两个月出去，我写信给Neyman教授，他给了我一个代理助理教授的工作，于是我就回来了，也就是那个时候我开始结识了David Blackwell。

我们那个时候一起开研讨会，他经常拿一些笔记来问我，“你会证明这个、这个和这个吗？”
有一次，他给的一个笔记上写着，“你能证明这个几乎收敛于下面的式子吗？”
我想了好些天，然后我觉得我可以做到，我觉得我已经有思路了，Shannon–Breiman–McMillan理论(Breiman, 1957)就是这么诞生的。和Dave Blackwell一起工作十分的鼓舞人心，他思考问题的方式真的非常了不起。

**Olshen: **Did he get youinto this whole business
of getting rich quickly on favorable games (Breiman,
1960)?

**Olshen: **是Dave Blackwell促使你进入博彩这个领域的吗？

**Breiman**：No. How that happened was pretty
much on my own. I was thinking about the Kelly
criterion.

**Breiman**：不是的，完全是因为我自己。我当时在思考凯利指数。

**Olshen: **Tell me about that.

**Olshen: **跟我讲讲怎么回事

**Breiman**：The Kelly criterion roughly says that
if youdo a certain kind of betting in a simple favorable
game, you can make your winnings go to infinity.
It was not almost surely, but that the expected
values would go to infinity. I thought about Kelly’s
results for a while. Then I realized that they might
lead to an optimum strategy for playing favorable
games.

And that’s what set me off. It was an accidental
thing. David was not involved. I don’t think that at
that point he was interested in gambling at all.

**Breiman**：凯利指数大概是说，如果你在一个有利的赌局中下赌注，你就可以赢得无穷多的钱。这也不是一定的，但是至少你的期望收益会是无穷的。我思考了一下凯利指数，然后我意识到他们其实是在采取最优策略来玩一局有利于自己的赌局。

这是我研究博彩的动机，完全是因为机缘巧合。David 没有参与进来，而且我觉得他应该对博彩业不感兴趣。

##SELLING ICE IN MEXICO(在墨西哥卖冰的那几年)
**Olshen: **It’s become clear already that you’ve
always had wide-ranging interests. One that’s fascinated
me, although I can’t pretend to understand
it yet, is the scheme for selling ice in Mexico. Can
you tell us about that?

**Olshen: **大家都知道你涉猎广泛，其中你在墨西哥卖冰那件事非常吸引我，但是我到现在也不太清楚，你能跟我讲讲那个吗？


**Breiman**：Right. Well, that’s actually due to a
couple of friends of mine who went down to Puerto
Vallarta in, I think, 1964 or ’5, the year after the
movie Night of the Iguana was made. This was when
Puerto Vallarta was just this little town and you had
to boil all of your water.



The other thing you had to be careful about was
if they put ice in your drink, don’t drink it. All
the local ice was made out of local water and was
swarming with bacteria. So my friends got this idea,
which also interested me, of getting a purified ice
machine down to Mexico and setting up a factory
for making purified ice.


I thought that was great and I said, “I’ll invest.”
What happened was, sure enough, they bought this
old ramshackled building called El Bucanero, the
bucaneer. They set up an ice plant in it and it was
a great success. And the ice was known throughout
Puerto Vallarta because they sucked the impurities
out of the middle so that the ice looked like
little donuts. And wherever youwent, if yousa w
a drink with little ice donuts in it, you knew you
were okay. After a while, it was so successful that
the mayor’s son decided to go into the purified ice
business. And sure enough, all our good customers
somehow started patronizing the mayor’s son’s ice
factory. And business dried out. But it was great
while it lasted.


**Breiman**：好的，实际上那是因为我的几个朋友在1964年或者1965年去了Puerto
Vallarta，具体年份我记不太清楚了，但是正好就是在Night of the Iguana(译者注：即《巫山风雨夜》，1964年上映的电影)这部电影上映之后的一年。
当时，Puerto
Vallarta还只是一个小城镇，你需要把你所有的水都煮沸才行。

另外一件你需要注意的事就是，如果有人在你的饮料中加了冰块，千万不要喝。
因为当地的冰块都是用当地的水做的，都充满了细菌。所以我的朋友就想出了一个点子：把
纯化的制冰机送到墨西哥，并且建立一个专门净化冰的工厂，这个想法同样让我觉得很感兴趣。

我觉得这很棒，我说，“我也要投资”。接下来他们买了一栋旧得几乎摇摇欲坠的建筑，叫做El Bucanero。
他们在这里建立起了冰厂，这是一个伟大的成功。然后这个厂净化出来的冰在Puerto Vallarta远近皆知，因为他们从冰的中间把杂质吸出来，所以看起来像一个甜甜圈。在Puerto Vallarta，无论你走到哪，你都可以看到一杯装有甜甜圈冰块的饮料。没多久，市长的儿子也决定开始做净化冰的生意。
果然，我们所有的顾客都跑去光顾市长儿子的生意了。所以我们的生意就黄了，但是净化冰这件事持续下来了，这非常伟大。


**Olshen: **Sounds good.

**Olshen: **听起来很棒

**Breiman**：Uh-huh.
**Breiman**：嗯，是的。

##THE UCLA YEARS（在UCLA的那几年）
**Olshen: **Now I’m going to change the subject
somewhat radically in order to segue into your
career at UCLA. And, in particular, I want to ask
yousomething that I’ve always wondered very
much about. And that is, how did you come to
write that wonderful book on probability (Breiman,
1968)?

**Olshen: **现在我要继续和你谈谈你在UCLA的职业生涯了，其实我是想问一些我一直非常想知道的事。
就是你怎么想到写probability (Breiman,1968)这本伟大的书的呢？

**Breiman**：When I went to UCLA I went because
I only had an acting appointment at Berkeley. And
those days the rule was pretty strict that if you were
a Berkeley Ph.D, you definitely couldn’t get your
first job at Berkeley. The next best thing looked to
me like UCLA.

So I walked in off the street and went to the Math
Department and said, “I would like to get a job.”
They got some letters about me, and I got a job. I
was there for about seven years, and I actually got
tenure fairly quickly. I was their only probabilist. So
I was teaching the graduate course in probability
theory.


I taught it about three or four years. And through
this whole teaching, I kept trying to figure out really
what was going on in probability theory—going on
in the sense of working on proofs until I really
understood or figured I could understand what
made the thing work.


Well, after about seven years at UCLA I said
to myself, “Look. Leo, you’re not cut out to be an
abstract mathematician. It’s been fun, but this isn’t
going to work for you.” So I resigned.

**Breiman**：我进入UCLA是因为，我在伯克利只有一份署理职位（译者注：署理职位一般不能超过6个月）。
那时的规则非常严格，如果你是一名伯克利的博士生，那么你铁定不能在伯克利找到你的第一份工作。那么接下来对我来说，去UCLA对似乎是最好的选择了。

所以我走在街上，并进入了数学系说，“我是来求职的”。他们收到了一些关于我的信件，然后我获得了一份工作。我在这儿一待就待了7年，实际上我很快就拿到了终身教授的职位。我是他们唯一的一名概率学家，所以
我教研究生概率论的课程，教了大概3到4年。

在整个教学过程中，我一直专注于概率论中的证明，直到搞懂内在的原理。

在UCLA待了大概7年之后，我对我自己说，“Leo,你天生就不是做抽象数学家的料，它虽然很有趣，但不是你该做的工作。”于是，我辞职了。

**Olshen: **Now, the book had been written by then?

**Olshen: **到那时，你已经写了这本书了吗？

**Breiman**：No.

**Breiman**：还没有

**Olshen: **Oh, it hadn’t?

**Olshen: **那个时候还没有吗？

**Breiman**：No.

**Breiman**：没有

**Olshen: ** Okay.

**Olshen: **好吧

**Breiman**：So I took out the money that I had put
into the retirement system, and I spent six months
doing nothing but writing the book. This was quite
a period of time because it took strong discipline to
write it. But again it was also a lot of fun, and it
was my first born.

Thus, I was tickled pink when SIAM approached
me just three or four years ago and said, “We want
to republish it in our classic series” (Breiman, 1968).

**Breiman**：所以我拿出了放在我退休金中的钱，然后我花了6个月时间专门来写这本书。
这是相当长的一段时间，因为写这本书需要很强的纪律性。
但是同时这个过程又充满了乐趣，它是我的处女作。

因此，当3、4年前SIAM([美国]工业及应用数学学会) 找到我说，“我们想在经典系列中再版你的书”时，我喜出望外。

**Olshen: **I’ve heard that youwent to Africa during
your time at UCLA.

**Olshen: **我听说你在UCLA工作期间，还去了非洲。

**Breiman**：Yes. This is the way it happened. I
wanted a sabbatical that would be really different.
So I went to UNESCO and I said, what have yougot
for me? I don’t want to teach in a university. And
UNESCO finally said, “We’ve got a place for youas
an educational statistician in Liberia.”

So I went to Liberia as an educational statistician
and what did they need? They needed to find out
how many kids they had in the schools. Why was
this a problem? They only had 50 miles of paved
road in the whole country. The schools were in the
rain forest jungles that were virtually inaccessible.
So we formed teams. We formed about 20 teams
to ride or paddle or walk all over the country, that
would go into little villages, call out the school children,
and just count them.

And so I lived there, I went on a number of these
treks into the backwoods, into tunnels in the rain
forest jungles that came to villages, and what was
amazing when I got there was that little children
would walk up to me and rub my skin to see if the
white would come off. That was an altogether fascinating
experience. I just loved the life in Africa and
I was sorry to leave.

**Breiman**：是的，确实是这样。我想要一个不同寻常的休假，所以我去了联合国教科文组织（UNESCO），然后我对他们说，“我不想在大学里教书，你们能给我什么其他的呢？”
最后联合国教科文组织告诉我，“我们有一个在利比亚（Liberia）当教育统计学家的机会给你。”

所以我就以一名教育统计学家的身份去了利比亚，那么他们需要什么呢？他们需要知道全国所有学校里的学生有多少人。为什么这会是个难题呢？因为整个国家只有50英里铺设好的道路。很多学校都在热带雨林丛林里，这些地方看起来根本没法进入。所以我们组了大概20支队伍，以骑车、划船和走路的方式走遍全国，深入到那些小村庄，把学校里的孩子们都叫出来，然后数一下人数。

所以我住在这里，艰苦跋涉去了很多边远地区，经过了热带雨林丛林中通往村庄的隧道。
很神奇的是，当我到达村庄的时候，小孩子们会过来搓我的皮肤，看看能不能把我皮肤上的白色搓下来。
那是一种非常有趣的体验，我几乎爱上了在非洲的生活，而且很遗憾要离开这里。

**Olshen: **Tell me now about your interactions
with Bill Meisel and about how you came to Technology
Services Corporation and your position
there.

**Olshen: **跟我讲讲你与Bill Meisel的合作，以及你是如何去TSC（技术服务公司）工作的

##AN INDEPENDENT CONSULTANT（一名独立的顾问）
**Breiman**：After I resigned from UCLA and I
used up my retirement money, it became pretty
clear I needed to do something to support myself.
So I asked around. And at the time, SDC—Systems
Development Corporation—which was an offshoot
of Rand, had a lot of Federal money for research
into freeway traffic.

I had written an article about how traffic
approached a Poisson process (Breiman, 1963) and
I knew someone who worked in traffic research.
And he said, “Sure. Youknow , why don’t you consult
for SDC.” So I consulted for them for about a
year on traffic. It was a lot of fun trying to figure
out what freeway traffic looked like and what its
statistical characteristics were.


Then that money dried up. So I was asking around
again, and Pete Payne, who was another consultant,
said, “Well, I think TSC” —this firm, Technology
Services Corporation— “is looking for consultants.”
So I went over there and Bill Meisel interviewed me.
We talked for a while. And he said, “Okay. You’re
on. When can youstart?” Bill was a very interesting
guy because he had been on the faculty of USC
Engineering.

**Breiman**：从UCLA辞职之后，我开始花我的退休金，很显然我需要干一些别的事业来养活我自己。
所以我四处询问，当时Rand旗下的SDC系统发展公司，有一大笔联邦资金用于研究高速公路交通。

我之前写过一篇关于交通流量服从泊松分布的文章(Breiman, 1963)，而且我认识一个在交通研究工作的人，他对我说，“你为什么不给SDC当咨询顾问呢？”所以接下来的一年我就给SDC当咨询顾问了。
研究高速公路交通流量数据长什么样以及它的统计特征是什么，是一件充满乐趣的事。

之后这笔联邦基金花完了，所以我又四处询问。另一位顾问Pete Payne告诉我，“我觉得TSC（技术服务公司）在招聘顾问。”所以我去TSC应聘，Bill Meisel面试了我。我们交谈了一会儿之后，他说，“OK，你被录用了，你什么时候可以开始呢？”Bill是一个十分有趣的人，因为他在USC工程系任教过。

**Olshen: **Did youknow him at Caltech?

**Olshen: **你之前在加州理工学院的时候就认识他吗？

**Breiman**：No. Not at all. And he had written a
book on pattern recognition. TSC was essentially a
radar house, a defense house, not a very big outfit,
maybe 50 or 60 technical people. And Bill was
running this sort of odds and ends division that he
decided to focus upon its becoming an Environmental
Division.


So I started consulting for the Environmental
Division. Between the two of us, we started writing
proposals to the EPA, to the California State
Air Pollution Board and so on. And we began winning
a lot of these proposals. I started doing a lot
of work that was EPA sponsored on air pollution,
water pollution and a variety of things of that sort.
A lot of the problems that we dealt with in those
days involved large amounts of data. For instance,
in one big project we had seven years of hourly and
daily data on over 450 variables relevant to air pollution.
We were trying to predict next day ozone
levels in the Los Angeles Basin.

Bill was very instrumental in getting me to think
about problems such as classification and regression.
For EPA, and many of the other places around
that were funding research, so many of the problems
were prediction and classification that this was
a hot area. And this was an area that Bill knew
something about. And he pushed me into this area.

**Breiman**：不，完全不认识。他还写过一本关于模式识别的书。
TSC基本上是一个雷达室，一个防御室，不是一个很大的公司，里面大概就是50到60个技术人员。
Bill当时管理这种零碎的部门，然后他决定集中精力让它成为一个环保部门。

所以我开始给环保部门当顾问，我们两个开始写提案给EPA（Environmental Protection Agency，美国环保署）和加利福尼亚州空气污染董事会等。我们开始赢得了很多赞成和支持。
我开始做一些EPA赞助的空气污染工作、水污染工作和各种各样诸如此类的工作。
我们遇到的很多问题都涉及到大量的数据，比如说在一个大项目里我们有7年每天每小时的空气污染数据，涉及到450个和空气污染相关的变量。
我们尝试预测洛杉矶盆地第二天的臭氧浓度。

Bill非常有助于我思考分类和回归的问题。对于EPA和周围很多其他资助研究的地方来说，大多数的问题
都是预测和分类，这是一个热门领域。Bill对于这个领域有所了解，他推动我进入这个领域。

Olshen: Smells like the beginnings of CARTR to
me. (Breiman et al., 1984).

**Olshen: **听起来这是CARTR(Breiman et al., 1984)的开始。


**Breiman**：Well, that’s right. We were working
on prediction problems like next day ozone in the
Los Angeles basin, carbon monoxide levels on freeways,
but also things such as could we recognize the
sender of handset Morse code—this was something
we were doing for the spook agencies—or could we
recognize from sonar returns whether the other
submarine was Russian or American? Or could we
recognize whether that battleship we were getting
radar returns from was Russian or American? So a
lot of this stuff was fascinating classification stuff.
At the same time, Bill went for almost every interesting
statistical request for proposals he could see.

For instance, one thing we did was a study of
delay in criminal courts in Colorado. We did some
other work for the criminal court system. I designed
some surveys and helped supervise surveys for the
EPA.

Thus, I was doing a whole bunch of things that
taught me about statistics—what you did with
statistics and data while youwere trying to solve
problems. Does that answer your question?

**Breiman**：是的，说的没错。我们在研究预测问题，比如说第二天洛杉矶盆地的臭氧浓度，
高速公路上的一氧化碳浓度，还有一些其他的研究，比如是否能识别出摩尔斯电码是从哪个手机上发送的（这是我们为间谍机构做的事）？或者我们能否通过声呐返回技术识别出潜水艇是俄罗斯的还是美国的？或者我们能否通过雷达回波技术识别出战舰是俄罗斯的还是美国的？很多这些问题都是迷人的分类问题。同时，
Bill对于几乎一切能看到的有趣的统计需求，他都提出建议方案。

比如，我们做的其中一项是Colorado刑事法庭的延迟研究，我们也为刑事法庭系统做了一些其它的工作。我也设计了一些调查来帮助EPA监督管理。

因此，从这一大堆工作中我对统计学有了更加深刻的认识，我学会了在尝试解决问题时，如何使用统计学和处理数据。这些回答了你的问题吗？