Tackling a complex problem (still relevant today) at an early age, getting great results and describing the solution clearly/concisely.
My master thesis was ~60 pages long, and was probably about 1/1000 as useful as this one.
Comparing your journey to others’ is pointless.
There is a very small number of truly gifted people who end up discovering things on their own at a young age. Most gifted people who discover something do so with the benefit of a mentor who can work with them to refine their talent into skill.
My masters thesis sucked largely because I tried to do it on my own and didn't even pick an advisor until I was almost done. At that point all they could do with my mess was to say "well, this is a decent descriptive paper and we need more descriptive papers in the field," and then give me proofreading comments. I didn't have a damn clue what I was doing, the end product was mediocre, and I didn't learn nearly as much as I could have. I'm not an outstanding talent by any means, but not seeking out mentorship in school is one of my only career-related regrets.
The fact of the matter is that that some people are deprived of mentorship, either through bad personal decision-making or through bad academic infrastructure. These people have a much harder road to expertise and success than the people who were mentored.
This was more to share with 'halflings that comparing one's achievements to someone else's is rarely something that bears much fruit.
My experience, mostly in grad school, was that anyone editing my work wanted more verbiage. If you only needed a short, one-sentence paragraph to say something, it just wasn’t accepted. There had to be more.
Jeff Dean is an uncommonly good communicator. But he also benefited from being allowed, perhaps even encouraged, to prioritize effective and concise communication.
Most people aren’t so lucky, and end up learning that this type of concision will not go over well. People presume you’re writing like a know-it-all, or that you didn’t do due diligence on prior work.
I _never_ got that feedback. My mentors all emphasized economy of language and nobody cared how "thick" my thesis was.
This is a pretty amusing story about verbiage.
Back in the old days, you would send a manuscript/research article to colleagues/friends by _snail-mail_ to get their feedback. You'd wait a month, and maybe they would mail a 'red-inked' copy of your manuscript back to you.
My Ph.D. advisor sent out a draft to a colleague who was famous for being harsh with the red-ink.
After a month, my advisor receives the manuscript in the mail.
* He turns to page 1. No red ink!
* He turns to page 2. STILL no red ink! [He must looove the paper]
* Keeps turning pages (no red ink!!).
* On page 10--in red ink--is written, "Start here."
So this puts the reviewer in a situation with misaligned incentives. They might prefer to tell you to prioritize concise communication, but believe the risk is high that such a thing will get vetoed by the committee for Dilberty reasons, and thus their feedback gets optimized for what the committee will superficially think.
When the committee is mostly attentive professors, this isn’t so bad and everybody is aligned on short, to-the-point style.
But my experience is that this is hardly true. Maybe one committee member will be an attentive technical authority, sometimes only your advisor. The others will be deans or directors of various sorts who view it as an administrative chore to even have to sign off, and probably farm that review out to grad students or adjuncts, who are far more likely to take a capricious point of view about e.g. heavy literature review or conclusion sections.
I wrote an undergrad thesis and I felt like my advisor and two readers cared about what I was saying and how I said it.
Same is true for my Ph.D. My committee members seemed to care deeply about the work (I think they just didn't want to be associated w/ crap research).
That shouldn't add too much. No more than a few pages. It would still concise but then also a scientific work.
How would you know the results are solid without consulting the scientific literature?
Papers that are entirely surveys or comparisons of different approaches can be excellent and would make good citations in any practical work.
This is not academic. What did reading this Master thesis teach me? That two approaches perform reasonably (by what standard?) with a size trade-off. That's an excellent start but also leaves open many questions: Why these two approaches? Are there reasons to expect they are better suited than other approaches in the literature? Were these results expected? Can I expect them to generalize? Do they paint a coherent picture on the performance of different designs in various contexts or are they surprising?
A lot of this is about generality of the knowledge gained. As a mere fact ("Two implementations of two algorithms that solve one problem perform slightly differently") it's not very interesting unless I have that exact specific problem myself. If I do, I would still need to find the paper. But if it is linked into a wider web of knowledge ("In paper [X] it was found that this algorithm performs well on tasks that have something in common with our problem, paper [Y] and [Z] suggest that we should expect a trade off for small sizes. Generally nothing is known about what should be algorithms well suited to the problem at hand.") it allows me to reason about situations.
Hence the desire to constantly look for novelty in academic work. Engineers don't necessarily care about novelty, they need to solve a problem at hand for practical reasons. Documenting what they've done, how it performed, and what they learned (if anything) is still important to write down for others who may want to solve similar problems.
I personally find the quest for novelty often reads like some kind of desperate need to justify the work or to get it funded. Solid work can stand on it's own even if there's nothing new about it, while mediocre work seems to stand so long as it's go some element of novelty.
If I've already decided what method I want to use to solve a problem, finding a well-done implementation and documentation on it is all I really want. If I don't know what solution to apply to a problem, a survey that documents the various approaches and makes some comparisons is what I want.
Also, I had to be submit 3 identical hard-bind copies of that bullshit report.
I had a friend who's advisor made them make everything longer the way you describe, theirs was in excess of 100 pages. (IIRC this advisor had suggested that while the guidelines say ~50 pages this was the bare minimum sufficient for a pass).
I guess it depends a lot on your advisor.
I value conciseness dearly, and prefer quality over quantity in scientific writing, i.e. I would accept incredibly short theses, if the content is sufficiently presented (reproducible and comprehensive), and most of all, contains a valuable contribution.
The reason I typically have to request "more verbiage" and an own section on the state of the art, is because I need to force my students to confront their sitcom ideas with the history of "what has been done before, and what the actual current problems are".
Unfortunately, the approaches of most students are neither new nor particularly interesting in this regard.
It's strange to expect an undergrad to do new and interesting work when they haven't even finished their basic education in the field. Solve problems that are easy but not important enough for professional academics, sure. Do an application of a standard idea in a specific environment (like porting an pp to Android), sure. But not new approaches to the field.
Scientific work should fulfill at least some standards, and IMHO this includes undergrad theses.
One thing most people don't get is that Dean is basically a computer scientist with expertise in compiler optimizations, and TF is basically an attempt at turning neural network speedups into problems related to compiler optimization.
I'd like to thank my undergrad university for hosting my undergrad thesis for 25 years with only 1-2 URL changes. Some interesting details include: Latex2Html held up, mostly, for 25 years and several URL changes. The underlying topic is still relevant (training the weight coefficients of a binary classifier to maximize performance) to my work today, even if I didn't understand gradient descent or softmax at the time.
Kudos to University of Minnesota (@UMNews) Honors Program. Earlier this year, I asked Prof. Vipin Kumar, my advisor for this work, if he still had a copy, since I had lost my copy. He didn't, but checked with the Honors Program and eventually got a very nice response saying: "Jeff and Prof. Kumar, Here is a pdf copy of the thesis in question. We digitized our physical library about 8-10 years ago and no longer keep hard copies of anything. Hope this is what you are looking for."
Lots of good work with neural networks was done back then:
A learning algorithm for Boltzmann machines
DH Ackley, GE Hinton, TJ Sejnowski - Cognitive science, 1985
Learning representations by back-propagating errors
DE Rumelhart, GE Hinton, RJ Williams - nature, 1986
Phoneme recognition using time-delay neural networks
A Waibel, T Hanazawa, G Hinton, K Shikano, KJ Lang - Readings
in speech recognition, 1990
The interest in NNs was ignited (in part) by this double volume collection of essays called "Parallel Distributed Processing" edited by Rumelhart and McClelland.
Dean even cites them. And, if you read the contributors, it contains many (though not all) of the heavy hitters.
Reading back on it, it will sound very familiar. All the amazing breakthroughs: object recognition, handwriting recognition etc all seemed to be there. But all that rapid progress just seemed to stop. There was this quantum leap and then you were back to grinding out for even 0.1% improvement.
For those who stuck through the second winter, things obviously paid off.
The intro essay is online:
Then when the data explosion started during the 00s, it laid the groundwork for the NN comeback.
My entire career has consisted of reimplementing bits and pieces of things I've previously built all the way back to high school and then reimplementing whatever was new on the previous round in the next one.
Also, not all intelligent American kids can or want to go to elite schools, even if they are academically qualified. In the U.S., you often hear stories of kids turning down really good schools for ones they felt were a better "fit" (financially, culturally, etc.). And unlike the rest of the world, elite colleges in the U.S. are often private and expensive. Despite need-blind admissions, not everyone can afford them without going into heavy debt. (many middle-class parents make just enough money for their kid to not qualify for substantial financial aid).
So kids go to schools they can afford.
One of my college professors (who attended Princeton and MIT) once told me that in his observation, the top 5 percentile students in (good) state schools aren't that different from the kids who went to Princeton or MIT. I didn't believe him at the time, but having worked with different folks over the years, my experience inclines me to believe that there's some truth in that observation.
Owing to its population and economy, the U.S. has a large enough talent pool that the top percentile students at large, well-funded state schools (of which UMN is an example) are plenty smart. If you were to meet the really smart top-5-percentile kids from such state colleges (I have), you'd have no doubt that many of them could have attended MIT or CMU.
To be sure, good colleges can give you a headstart in life -- but it's what you do with that advantage that counts.
Examples of smart computer folk who went to decent, but non-elite schools for undergrad:
JJ Allaire (ColdFusion, Rstudio, etc.), Macalester College
Ward Cunningham (Wikis), Purdue
Rich Hickey (Clojure), SUNY Empire State (though he did go to Berklee College of Music)
John Carmack (Doom, Quake), U. Missouri Kansas City
Sergey Brin (Google), U. Maryland College Park (before Stanford)
Larry Page (Google), U. Michigan (before Stanford)
Dave Cutler (VMS, Windows NT), Olivet College
Bram Cohen (BitTorrent), U at Buffalo
Ryan Dahl (Node.js), UCSD, then U Rochester
Larry Wall (Perl), Seattle Pacific U (before Berkeley)
Alan Kay (Smalltalk, windowing GUIs), U Colorado, then U Utah.
Apologies for not making myself clear enough :(
> To be sure, good colleges can give you a headstart in life -- but it's what you do with that advantage that counts.
I just graduated undergrad from a state school (Rank #49 in CS) but I'm still pretty skeptical of this fact.
It's more common to hear large public Engineering universities & technical institutes.
And as another mentioned, Google is so big that most hires can't be from a few small schools.
Bonus question: What if we change our constraints to only account for 3-sigma people? Does our conclusion change?
This speaks to their lack of confidence more than their capability. Perhaps that is one of the advantages a good school, a good peer group, or a good network can confer: the confidence to aim higher.
People don't think they're good enough... which may be true, but no one can truly know until they try. Self-limiting thinking is particularly prevalent in rust-belt cities and regions where knowledge or achievement is not prized, so people in knowledge-intensive fields have no models to emulate.
And sometimes when they try (it's not unusual for graduates of lower ranked universities to send out 300+ resumes only to get single digit responses), they get demoralized when they don't succeed on their first few tries, when in fact there's more than one path in life -- if one doesn't have natural advantages, one might have to embrace the more circuitous path(s). This can mean joining a startup, going to a better grad school than one's undergrad, moving to a better city to
upgrade one's peer groups (this is more important than most people think ), etc.
Life can surprise you if you keep trying and pivoting (ugh cliche, but there it is). There's an element of randomness and stochasticity in a free market, and I've seen enough counterexamples to distrust a static conception of how things "should be". (except for some stodgy areas like investment banking that only hire from certain schools; but even then there are backdoors)
* You're a new graduate, and the hardest hurdle you have to overcome is to get in the door. If you manage to do that and are able to prove yourself, your undergrad degree will become less and less and important. If you google Fortune 500 company CEOs, especially in non-tech companies, (you can do this exercise for yourself) you will learn that many of them went to non-elite schools for undergrad. For all its elite colleges, America is not really an academic-technocratic society (unlike countries like Germany where most CEOs have Ph.D.s). There are elements of William James' pragmatic philosophy that still influence the thinking in this country -- getting results is more important than academic knowledge.
I can say that CSE was very selective when I was there, and getting into upper division was even harder. But overall I don't think acceptance rate is a very useful statistic because program size affects it so much.