↑ 点击蓝字 关注极市平台

来源丨管理学季刊

编辑丨极市平台

极市导读

 

为帮助博士生们顺利毕业，哈佛医学院的Matt Might教授总结出了读博失败的十大原因，包括学习过泛、目标过高、完美主义、行动拖延等，可谓鞭辟入里。

博士难毕，全球皆如此。差不多每个学校都有1/3到一半的博士研究生拿不到学位。读博失败不仅本人难受，导师也不好过。为了帮助自己的博士生们顺利毕业，Matt Might教授（计算机科学家、生物学家、教育家），总结并发表了著名的“10 Easy Ways to Fail a Ph.D.(读博失败的10大常见原因)”，可谓是鞭辟入里、针针见血。其中很多观点，比如对“完美主义”的看法，和我们之前分享的施一公教授和诺奖得主的四条箴言中的观点不谋而合。Matt Might教授表示，这10点，不管什么专业的博士生，可以说是大家的都容易犯的通病。分享给大家。

Matthew Might

计算机科学家，生物学家，教育家。2017年起任阿拉巴马大学伯明翰分校（UAB）Hugh Kaul精准医学研究所所长。2001年获得佐治亚理工学院的学士学位，2007年获得了计算机科学的博士学位。2008-2017年，任职犹他大学，一直担任计算机科学和药物化学教授；2017年，离开犹他大学，成为哈佛医学院生物医学信息学客座教授。

01 过度在意成绩或课程作业

实际上，没人会在意你的分数是多少。这里有一个计算GPA的通用的公式：

最优GPA=最低要求的GPA + ε

更高的分数意味着本应该用于研究的时间被浪费在了课堂上。教授们甚至可能会对4.0的分数感到惊讶。在头两年里，学生应该选择一个感兴趣的研究领域，并根据研究兴趣跟随一位导师，然后通过大量论文阅读后，着手尝试小型的、探索性的研究项目，而在课程上花费过多的时间只会让你离真正的目标越来越远。

No one cares about grades in grad school.There's a simple formula for the optimal GPA in grad school:

Optimal GPA = Minimum Required GPA + ε

Anything higher implies time that could have been spent on research was wasted on classes. Advisors might even raise an eyebrow at a 4.0. During the first two years, students need to find an advisor, pick a research area, read a lot of papers and try small, exploratory research projects. Spending too much time on coursework distracts from these objectives.

02 什么都想学

有同学读博因为他们就是爱学习、想学习。读博的过程的确需要大量的积累与学习，但有一点千万要搞明白：博士生的学习是要将精力集中在毕业论文方向上进行有针对性的学习，是以形成论文为最终目的的。过多的选修或者旁听一些与你研究方向无关的课程，可以说，基本上都是在浪费时间，完全没必要。

到第三学年末，一个合格的博士研究生应该已经阅读了50到150篇与你研究方向最新成果相关的学术论文。当然，有的学生在相关研究上走得太远了，专研了一大堆他们将来打算做却还没有着手的研究领域的学术资料。导师们会对这些“志向远大”的学生失去耐心——这些学生醉心于为人类知识做出重大贡献，而不是脚踏实地的完成既定的目标。

我当年读博的时候，为了实现对自我潜力的挖掘，就患上了“什么都想学”的毛病。在我读博的头两年，学校里开什么课我都跑去上：阿拉伯语、语言学、经济学、物理、数学、心理学。在计算机科学方面，我耗了不少时间上了很多与我的研究无关的课程。所有这些“启蒙”的代价就是——我延期一年才获得了博士学位。

我之所以走了弯路，主要是因为当年我瞎折腾的时候，我正在担任助教，对我的导师来说，这笔助教费花的挺值的。

Some students go to Ph.D. school because they want to learn.Let there be no mistake: Ph.D. school involves a lot of learning.But, it requires focused learning directed toward an eventual thesis.Taking (or sitting in on)non-required classes outside one's focus is almost always a waste of time, and it's always unnecessary.

By the end of the third year, a typical Ph.D. student needs to have read about 50 to 150 papers to defend the novelty of a proposed thesis.Of course, some students go too far with the related work search, reading so much about their intended area of research that they never start that research.Advisors will lose patience with "eternal" students that aren't focused on the goal--making a small but significant contribution to human knowledge.

In the interest of personal disclosure, I suffered from the "want to learn everything" bug when I got to Ph.D. school.I took classes all over campus for my first two years: Arabic, linguistics, economics, physics, math and even philosophy. In computer science, I took lots of classes in areas that had nothing to do with my research.The price of all this "enlightenment" was an extra year on my Ph.D.

I only got away with this detour because while I was doing all that, I was a TA, which meant I wasn't wasting my advisor's grant funding.

03 完美主义

在学术界，完美主义可以说是一种悲剧性的折磨，因为它往往对最聪明的那群人打击最大。你永远无法做到完美，最多无限接近。有同学常常对一篇研究论文的打磨远远超过了收益递减点，他们不断追求完美，永远不停的润色、打磨。

想在规划出完美的论文结构之后才能开始下笔的同学，永远下不去笔。

对于刚开始写论文或在写作中遇到问题的同学，我的建议是：写论文应该是一个反复修改、不断琢磨的过程。

从研究的提纲和一些粗略的笔记开始着手进行，然后在此基础上不断修改、迭代、打磨，然后不断重复这一过程。当在这个过程中文章的变化越来越小时，它就到达了收益递减点。在这一点上，它只需要再通过一两次修改即可完工了。

在论文的写作中“足够好”远远胜过“完美”。

Perfectionism is a tragic affliction in academia, since it tends to hit the brightest the hardest.Perfection cannot be attained. It is approached in the limit.Students that polish a research paper well past the point of diminishing returns, expecting to hit perfection, will never stop polishing.Students that can't begin to write until they have the perfect structure of the paper mapped out will never get started.

For students with problems starting on a paper or dissertation, my advice is that writing a paper should be an iterative process: start with an outline and some rough notes; take a pass over the paper and improve it a little; rinse; repeat. When the paper changes little with each pass, it's at diminishing returns. One or two more passes over the paper are all it needs at that point.

"Good enough" is better than "perfect".

04 拖延症

慢性完美主义者往往也是拖延症患者。那些“志向远大”的热爱学习、而不是搞研究的学生往往都有拖延症。读博士似乎对所有拖延症患者都有极强地吸引力，不幸的是，它同时也像是一个筛子，淘汰了那些效率低下的研究生。拖延症患者应该看看我写的关于提高效率的建议。

Chronic perfectionists also tend to be procrastinators. So do eternal students with a drive to learn instead of research.Ph.D. school seems to be a magnet for every kind of procrastinator.Unfortunately, it is also a sieve that weeds out the unproductive.Procrastinators should check out my tips for boosting productivity.

05 “独立”得过早或过晚

是被指导还是给建议，读博时，导师和学生的角色会随着课题研究的不断深入发生转变。研究的早期，指导老师应该手把手进行指导，帮助学生选定主题并助其成早期的论文。在研究的后期，学生对于课题的了解应当远胜于导师。一旦发生这种转变，学生应该开始走向“独立”，开始学会自己选择主题，并开始着手撰写论文。即使她的导师坚持让她做别的事情，她也需要学会“独立”。

关键在于把握好时机。

在学生还没搞清楚如何选好题、开好题之前让她独立去选题，最后不光她交上来的论文是废纸一堆，导师也会被气个半死。

另一方面来说，如果坚持要到某一个点以后才放手让学生去做选题，考虑到导师投入的时间和产出比，以及来之不易的资助，这会让导师压力倍增，精疲力竭。

导师们期望毕业年级的博士研究生们能达到讲师（proto-professor）的水平——对自己研究领域的具有挑战性的问题非常熟悉。他们应该有能力去做好选题，并以合适的篇幅和视角去解答研究课题。

The advisor-advisee dynamic needs to shift over the course of a degree. Early on, the advisor should be hands on, doling out specitic topics and helping to craft early papers.Toward the end, the student should know more than the advisor about her topic. Once the inversion happens, she needs to "go rogue" and start choosing the topics to investigate and initiating the paper write-ups. She needs to do so even if her advisor is insisting she do something else.

The trick is getting the timing right

Going rogue before the student knows how to choose good topics and write well will end in wasted paper submissions and a grumpy advisor.On the other hand, continuing to act only when ordered to act past a certain point will strain an advisor that expects to start seeing a "return" on an investment of time and hard-won grant money.

Advisors expect near-terminal Ph.D. students to be proto-professors with intimate knowledge of the challenges in their field. They should be capable of selecting and attacking research problems of appropriate size and scope.

06 把读博当成是在上学或工作

读博士既不是在上学，也不属于工作。

读博是一种不断修行的体验过程，也是一项值得别人羡慕、令人嫉妒的经历。解决难题、完成一篇漂亮的论文获得同行的好评需要日以继夜的苦思冥想。

你要以宗教般的狂热与虔诚看完所有相关研究的文献资料，博士研究生涯甚至还意味着贫穷和服从，毕业时你才可能得到一件期待的博士服和一顶博士帽。

那些把读博当做是朝九晚五在公司上班的研究生们，得花上至少7年才能获得学位，甚至只能拿到准博士（ABD）。

Ph.D. school is neither school nor work.Ph.D. School is a monastic experience. And, a jealous hobby.Solving problems and writing up papers well enough to pass peer review demands contemplative labor on days, nights and weekends.

Reading through all of the related work takes biblical levels of devotion.Ph.D. school even comes with built-in vows of poverty and obedience.The end brings an ecclesiastical robe and a clerical hood.Students that treat Ph.D. school like a 9-5 endeavor are the ones that take 7+ years to finish, or end up ABD.

07 不把委员会当回事

有的研究生忘记了学位委员会有权取消他们的学位。

对于学生们来说，在最后几年里和学位委员会的委员们保持联系非常重要。委员们需要知道你在研究什么。而且这些委员会成员可不像你导师那样对你耳提命面，你会很容易忘记他告诉你的建议。

尽管不常发生，但是我的确曾见过一位学位委员会的老师和答辩人因为意见不一致大吵起来，他们在用于评估实验的指标上有分歧。这位老师曾在这位学生请教他的时候告诫过他指标选取有问题，但这位研究生忽视了他的建议。不过，这位学生还是挺走运的——他只是被延期毕业了一个学期。我知道还有一位研究生，不是在答辩期间，而是在他提交论文初稿的时候就无视学位委员会老师的建议，结果没有获得答辩资格。学位委员会告诉他——整个初稿推到重来。于是，他整整花了十年才完成他的博士学业。

Some Ph.D. students forget that a committee has to sign off on their Ph.D. It's important for students to maintain contact with committee members in the latter years of a Ph.D. They need to know what a student is doing.It's also easy to forget advice from a committee member since they're not an everyday presence like an advisor.Committee members, however, rarely forget the advice they give.

It doesn't usually happen, but I've seen a shouting match between a committee member and a defender where they disagreed over the metrics used for evaluation of an experiment. This committee member warned the student at his proposal about his choice of metrics.He ignored that warning.He was lucky: it added only one more semester to his Ph.D.

Another student I knew in grad school was told not to defend, based on the draft of his dissertation. He overruled his committee's advice, and failed his defense. He was told to scrap his entire dissertation and start over. It took him over ten years to finish his Ph.D.

08 目标定的太低

有的研究生以最差的学生作参照，指望花最少的力气拿到博士学位。这样态度可想而知：将来他们肯定拿不到教职。而且，十有八九是毕业答辩要挂了的。

这些人挂掉基本上都是因为不幸选题不好或者计划做得太随性。

目标过低将导致你无法应对任何的不确定性。而研究恰恰是充满了不确定性的。

Some students look at the weakest student to get a Ph.D. in their department and aim for that. This attitude guarantees that no professorship will be waiting for them. And, it all but promises failure.The weakest Ph.D. to escape was probably repeatedly unlucky with research topics, and had to settle for a contingency plan.Aiming low leaves no room for uncertainty. And, research is always uncertain.

09 目标定的过高

对博士研究生来说，拿下博士学位那就是读博期间的主要目标，的确如此，但是，这不是终极目标。恰恰相反，这是你学术生涯的开端，是入场券。

获得博士学位不一定要能够治愈癌症或实现冷聚变。相信还有少数化学家记得爱因斯坦的博士学位论文，爱因斯坦的博士论文就是一个原则性的计算，旨在估计阿伏伽德罗常数。并且他还弄错了，算成了其3倍，但他还是拿到了最终的博士学位。

博士论文对人类知识的贡献虽小但却意义重大。它的影响在于其是学生一生所要从事研究的目标。获得博士学位所产生巨大影响就如同你第一次开枪就击中靶心那样。一旦你知道如何射击，你就可以一直射击，直到你击中它为止。

而有了博士学位，你就如同有了终身的弹药供应。

有的导师会给你一个可以参考的研究课题清单，如果他们同意的话，你可以在里面挑一个感兴趣的而且最容易的试试。

但是要记住：你的博士学位是什么并不重要，重要的是你能拿到一个。

博士生涯对思维的训练远远胜过你选择了一个什么样的课题。

A Ph.D. seems like a major undertaking from the perspective of the student. It is. But, it is not the final undertaking. It's the start of a scientific career.A Ph.D. does not have to cure cancer or enable cold fusion. At best a handful of chemists remember what Einstein's Ph.D. was in. Einstein's Ph.D. dissertation was a principled calculation meant to estimate Avogadro's number. He got it wrong. By a factor of 3.He still got a Ph.D.

A Ph.D. is a small but significant contribution to human knowledge.Impact is something students should aim for over a lifetime of research.Making a big impact with a Ph.D. is about as likely as hitting a bullseye the very first time you've fired a gun.

Once you know how to shoot, you can keep shooting until you hit it.Plus, with a Ph.D., you get a lifetime supply of ammo.Some advisors can give you a list of potential research topics. If they can, pick the topic that's easiest to do but which still retains your interest.It does not matter at all what you get your Ph.D. in.A ll that matters is that you get one. It's the training that counts--not the topic.

10

误解了真正的里程碑

  Miss the real milestones   

大部分学校都要求学生完成课程设计、课程合格、开题报告、论文答辩和毕业论文。这些都是写在纸面上的要求。

实际上，真正的里程碑是能发表三篇像样的、和你研究课题（大致）相关的好论文。

课程设计和课程合格的机制设置是为了能挽回在招生时的失误。一位学生如果发表了论文就证明了她每科顺利过关是货真价实的。

一旦一位学生发表了两篇好论文，而且她也能使答辩委员会相信她即将发表第三篇，那她就可以开题了。

假如一位学生发表了三篇学术论文，鉴于她可以持续的发表具有一定价值的报告并获得同行的认同，那么她就可以满怀信心的参加答辩了。如果她的这些文章都围绕一个主题，又有论题，还把她发表过的东西都装订成册，那她连毕业论文都弄好了。

我幻想去买一个工业订书机，把发表的三份学术期刊装订在一起，并把它称作“答辩天书”（The Dissertator）。

当然，三篇发表论文远不足以获得教职——哪怕是在一个烂学校。不过，对拿到博士学位来说，够了！

Most schools require coursework, qualifiers, thesis proposal, thesis defense and dissertation. These are the requirements on paper. In practice, the real milestones are three good publications connected by a (perhaps loosely) unified theme.Coursework and qualifiers are meant to undo admissions mistakes. A student that has published by the time she takes her qualifiers is not a mistake.

Once a student has two good publications, if she convinces her committee that she can extrapolate a third, she has a thesis proposal.Once a student has three publications, she has defended, with reasonable confidence, that she can repeatedly conduct research of sufficient quality to meet the standards of peer review. If she draws a unifying theme, she has a thesis, and if she staples her publications together, she has a dissertation.

I fantasize about buying an industrial-grade stapler capable of punching through three journal papers and calling it The Dissertator.Of course, three publications is nowhere near enough to get a professorship--even at a crappy school. But, it's about enough to get a Ph.D.

原文链接：http://matt.might.net/articles/ways-to-fail-a-phd/

推荐阅读

• 如何规划及践行博士生涯：我的个人经验
  

• 没有顶会的CV/NLP方向的博士生毕业出路在哪里？
  

• 21页博士生存指南，佐治亚理工助理教授亲自撰写，有志读研的看过来
  

添加极市小助手微信（ID : cvmart2），备注：姓名-学校/公司-研究方向-城市（如：小极-北大-目标检测-深圳），即可申请加入极市目标检测/图像分割/工业检测/人脸/医学影像/3D/SLAM/自动驾驶/超分辨率/姿态估计/ReID/GAN/图像增强/OCR/视频理解等技术交流群：每月大咖直播分享、真实项目需求对接、求职内推、算法竞赛、干货资讯汇总、与 10000+来自港科大、北大、清华、中科院、CMU、腾讯、百度等名校名企视觉开发者互动交流~

△长按添加极市小助手

△长按关注极市平台，获取最新CV干货

觉得有用麻烦给个在看啦~