Thread

Doing a Machine Learning PhD While Working in Japan

145 points · 44 comments · pwim

Loading the complete thread in the background. This saved snapshot is available now. Refresh

sashank_1509 · 1790803128 · focus · HN ↗
I would worry more about AI automating your PhD than whether you did it in Japan or EU
1. suddenlybananas · 1790803316 · focus · HN ↗
  What's the point of this comment? Did you even read the article?
  1. rrr_oh_man · 1790805351 · focus · HN ↗
    psychosis
    msdz · 1790806425 · focus · HN ↗
    No, you don’t get it. The until now completely un-recursive not-self-improving training runs will automate themselves away any day now, and ML PhD’s will become worthless!!
    sashank_1509 · 1790878182 · focus · HN ↗
    They already are imo, 50k submissions to ICLR, no ones doing any research. Most computer vision research invalidated by GPT Astra
    suddenlybananas · 1790884267 · focus · HN ↗
    Do you think every submission is meaningful, insightful or useful?
2. gregjw · 1790814196 · focus · HN ↗
  Surely considering the field he got the PhD in, hes in the safest field for job security regarding this.
  1. sashank_1509 · 1790873087 · focus · HN ↗
    It’s the opposite, ML is most dependent on excess compute.
rrr_oh_man · 1790805439 · focus · HN ↗
This sounds SO.MUCH. like the type of hedging academic writing I despise
```
  Doing a PhD is difficult and isolating, which is exacerbated by being an international student in an unfamiliar country. And so, while I overall very much enjoyed my time there and have freely recommended it to many people, the decision to pursue a PhD in Japan should always be considered carefully.
```
(Watch <a href="https://news.ycombinator.com/item?id=49879250">https://news.ycombinator.com/item?id=49879250 why it's bad)
1. LtWorf · 1790806709 · focus · HN ↗
  If you can't detail why it's bad in any shorter form than a 90 minutes video, perhaps it's not that bad.
  1. monkpit · 1790813952 · focus · HN ↗
    Agentless passive, hedging, zombie nouns. It’s the type of writing that’s meant to meet a word count.
    amemi · 1790818596 · focus · HN ↗
    But that does not seem to be related to academics, as GP called it.
    Perhaps it is the contrast of "freely recommending" and "always should be considered carefully"?
    Or taking a larger leap: the awkwardness of saying how hard (isolating) it is to do a PhD in an unfamiliar country, then saying that you enjoyed it, but others should consider it carefully?
  2. r_lee · 1790819897 · focus · HN ↗
    I would have to agree with the guy, I mean:
    "And so, while I overall very much enjoyed my time there and have freely recommended it to many people, the decision to pursue a PhD in Japan should always be considered carefully."
    literally sounds like "as an AI model..." type of disclaimer or padding
    like, who isn't going to carefully consider on how/where to do their PhD?
    LtWorf · 1790837560 · focus · HN ↗
    I agree with you. But I will point out you wrote it without needing to spam your own 90 minutes video.
    rrr_oh_man · 1790842287 · focus · HN ↗
    Eat a snickers :)
tkgally · 1790810650 · focus · HN ↗
[delayed]
1. mcyc · 1790811150 · focus · HN ↗
  Thanks, that's good to know, and I should have been a bit more precise.
  What field were you supervising?
  3 years in CS is, imo, a little tight but not bad. But I always wondered how people in other fields (stem and otherwise) did it. CS simply doesn't have to deal with the extraneous factors that come with, e.g., biology or climatology.
  1. tkgally · 1790813784 · focus · HN ↗
    I was affiliated with a graduate department in language and information sciences, and the fields of the faculty and students ranged from literature, history, and philosophy to theoretical linguistics and natural language processing. I don’t recall any students doing hardcore CS, but it wouldn’t have been inconceivable, especially if there was an NLP connection. But the academic culture of the department leaned more toward the humanities and social sciences than to science and engineering.
    One of the factors making the finish-in-three-years target very difficult to meet is that the thesis committees would include faculty from different fields and backgrounds, often with differing expectations for what a Ph.D. thesis should include. It was not unusual for a committee to ask a student to make a variety of substantial changes and additions to a thesis draft before submitting. That revision process could take six months or longer; because of the schedule for graduate coursework and defenses, that alone would push the student beyond the three-year target.
    Add in all of the other stumbling blocks that might come up in graduate school—family issues, financial problems, an advisor who retires or gets ill, etc.—and three years was no longer a realistic goal in that department.
mcyc · 1790810690 · focus · HN ↗
I am the author of this. Happy to discuss!
One thing to note is that, after I submitted the final draft of this article, the Japanese government ended the MEXT university track or is at least planning to. There was a memo, but the status is not clear.
Anyway, I had a great time doing my PhD this way, and generally I recommend that people work before a PhD and try to work during it (if you are in CS). It gives you a lot of freedom and, at least for me, it is very helpful to have more than one pipeline of work at a time so that when one slows down (as it inevitably does in research), you can draw on the other.
1. sbinnee · 1790815928 · focus · HN ↗
  Congrats! I am surprised that you first chose Korea for PhD. As myself being Korean, I hear a lot of horrendous stories about PhD students in Korea. They tend to spend for more than +8 years. I know a few people who went through their PhDs and certainly it took +8 years for all of them. I did my PhD by the way in France and it took me 4 years.
  Did you know that it could take so long when you first started your PhD in Korea? Could this be one of your reasons to move to Japan?
  1. sstim · 1790816888 · focus · HN ↗
    FWIW, the "extended" Korean PhD is an artifact of grant projects that need people to do the busy work (usually reports), and it seemed to be uncommon to have foreign students thrown into that pool. Scholarship max durations are also a limiting factor.
  2. mcyc · 1790820425 · focus · HN ↗
    Thanks!
    My decision to start my PhD in Korea was under very different circumstances. I was studying a totally different field (automata theory, so closer to mathematics than ML) where longer PhDs were normal, I was used to the US system, I wasn't working or married, etc. I expected it to take ~6 years when I started.
    I basically had to choose between living in the Canadian countryside, Helsinki, or Seoul in order to be in a lab studying what I wanted. I was already pretty into Korean culture (I minored in Korean in college, studied abroad there, and was a serious StarCraft player) so that choice was easy!
    When I started looking again, I knew I wanted a shorter PhD (especially since I had a masters and industry + industry research experience), so Japan was pretty attractive.
    I will admit that I think 3 years is a little short to make substantive progress outside of machine learning, so unless you come in with a lot of research experience and a good idea of what you want to work on, I think a longer PhD would be beneficial, but the opportunity cost is so high that it really doesn't seem like a great option on paper.
    fukaiall · 1790848091 · focus · HN ↗
    As a Starcraft fan, if you’ve not heard of yet, I highly recommend the Youtuber Kim Seong-Hyun (Alphago)
2. advael · 1790829693 · focus · HN ↗
  Do you see any path to doing this that doesn't go through working big tech? I'd love to finish a PhD and have at least a few ideas of what I'd do at this stage for it, but few companies are willing to sponsor this kind of thing, and many universities won't even allow you to work while in their program
3. rstuart4133 · 1790831430 · focus · HN ↗
  [delayed]
  1. rcxdude · 1790843085 · focus · HN ↗
    Predictive texting/typing was basically the use for language models before LLMs. Not all language models are LLMs.
  2. froh · 1790846495 · focus · HN ↗
    swipe/glide typing?
4. gkanai · 1790831929 · focus · HN ↗
  Thank you for sharing.
  "In essentially every other category, my quality of life has degraded significantly since returning home. "
  This is the challenge. The market in the US, the salaries, etc. are significantly better/higher, but the quality of life is often worse.
  1. Grombobulous · 1790859127 · focus · HN ↗
    And that’s quite notable since the suburbs of Silicon Valley is one of the best places to live in the entire country.
    iamEAP · 1790860222 · focus · HN ↗
    Though the author enumerated the some of the subjective things that rolled up into “QoL” and SV suburbs objectively don’t meet them (eg walkability).
    Grombobulous · 1790862392 · focus · HN ↗
    Right, so if you then downgrade from the Silicon Valley suburbs with their excellent weather, strong economy, ample cultural and education infrastructure and head to some other less attractive suburb in the United States you’ve gone even further down the lower quality of life rabbit hole.
    square_usual · 1790865641 · focus · HN ↗
    Not universally. The author wants to live in a city, so of course they're not going to enjoy living in a suburb. It sounds like they'd be better off in an East-cost metropolis like New York instead.
    Grombobulous · 1790906149 · focus · HN ↗
    They could also live in San Francisco proper, too, without being far from family.
5. mastermage · 1790835512 · focus · HN ↗
  Hi there for reference you said self funded PhDs are common in Japan. Would you hazard a guess what it would cost to fund your own PhD. I have lived in Japan before during a half year of language school, and still want to at one point in the not so far future do my PhD. And Japan would be one of the potential places on the list.
  1. mcyc · 1790838647 · focus · HN ↗
    Hmm, the tuition is extremely low relative to the US. On the order or $4-5k USD/year. I am not sure how many additional fees (like facilities, etc.) there are, but I guess not many. Being self funded was common for Japanese students. AFAIK no international students that I knew were self funded, up to the tail of their graduate program when their MEXT ran out.
    Now that I think about it, one thing I am not sure about is if self-funded students' travel/conference fees were covered. I would guess they are covered through a grant and it is just the tuition that is self funded, but I do not know for certain. Going to >1 international conference a year would be far more than the tuition.
    So, I would guess it would cost ~10k/year to be fully self funded. You can recoup some of that cost working (on a student visa this is limited though), and your lab may also have an RA stipend (but this is both very low [like $20/hr] and counts towards your 28 hour/week limit).
    My advice though is to never do a self-funded PhD. Being a PhD "student" is a misnomer. You are a research employee and should be compensated as such.
6. woweink · 1790862974 · focus · HN ↗
  [dead]
7. johnwalkr · 1790948926 · focus · HN ↗
  I had no idea they were ending this program. I had a MEXT scholarship for a PhD in aerospace engineering. In my case I tried the recommendation via school first and didn't get it even though I did have a letter from a Professor. By coincidence a year later I had an internship in Japan with another Professor and with his support I got into the program via the embassy route.
  Your experience mostly tracks with mine, but in my case there was a lot of interaction between Japanese and foreign students, and international journal and conference submissions were preferred. Many courses were offered in English and those that weren't, I could do homework and tests in English. Thesis had to be in English, even for Japanese students. It was a goal of the program to also help Japanese students get comfortable with international collaboration and working in English.
  In my case I stayed in Japan for a decade and it was overall a very good experience.
  One note about working as a student in Japan. You can work up to 28 hours a week without limitations as you noted. You can even work remotely for a foreign company, which is unique because if you have a work visa and freelance in Japan either as your main job, or a side job, your main client is supposed to be Japanese. This is a slightly grey area, your taxes are to be paid in Japan, and the company will be "operating in Japan" so be very careful about doing this by the books.
drnick1 · 1790818955 · focus · HN ↗
While I would certainly consider jobs abroad, I think that not getting your PhD in the US is a huge risk, especially as an American who might eventually want to return to the States. American PhDs basically trade a substantial premium and are employable in academia worldwide, while the converse is exceedingly rare. Most foreign academics working in the US got their PhDs here.
1. autuni · 1790838780 · focus · HN ↗
  That's not true. Academics working in the US are mostly lured in by high paying job offers, after they have already established themselves as researchers.
  The focus on name-recognition in the article was weird to me, but then again I am not from the US so I don't know how it is handled there. From the academic world I know, as soon as you're beyond undergrad level the university does not really matter as much anymore, because you should have published research and other contributions by the time you apply for anything at that point, which is what is being looked at. What's the quantity and quality of your (peer reviewed) published research, was it published in good journals or good conferences, etc.
2. cycomanic · 1790862635 · focus · HN ↗
  I don't know what field you are talking about, but your statement is simply not true for the fields that I'm familiar with (physics and EE). I know plenty of foreign PhDs who went to the US for academic or research positions, at postdoc or faculty level.
kimurakaera · 1790822902 · focus · HN ↗
I did my PhD at Titech (Suzukakedai).
The 3-year limit is not a hard limit. It is common to have an extension. It may also depend on the department. Most foreign students drop after 3 years due to a lack of funding.
Compared to the research culture in the UK/US, I think Japan is much better. Yes, you may be forced to overwork and sleep at the lab, but the guidance and supervision are good.
The only thing I hated was that COVID hit right before the graduation ceremony. All we had was an online ceremony.
1. Onavo · 1790824489 · focus · HN ↗
  > Yes, you may be forced to overwork and sleep at the lab, but the guidance and supervision are good.
  I think this is very much YMMV. Yes it's helpful if you are doing research in a narrow domain that benefits from hands-on guidance. But if you have your own ideas, a laissez-faire laboratory head and a shit ton of funding goes a lot further than riding on your professor's coat tails.
2. mcyc · 1790829371 · focus · HN ↗
  Yeah, it is definitely a harder limit for international students for the reason you stated (but this article is geared towards foreigners after all).
  What field were you in? I am surprised to hear that you think Japan's lab culture is much better than the US. I do think it varies a lot by lab, but overall the internal collaboration and guidance (if not from the PI, from senior members) was much better in American labs than in Japanese labs in my experience.
musicale · 1790828350 · focus · HN ↗
> I recently completed my PhD in Computer Science at Tokyo Institute of Technology (now the Institute of Science Tokyo)
They never knew why their collegiate apparel business was so successful...
1. motionlessveloc · 1790834740 · focus · HN ↗
  Which is why they go (went) by "Titech".
esrh · 1790834506 · focus · HN ↗
I just finished my MS doing ML research at tokyo tech (and also went to georgia tech!!), and I don't think I would honestly recommend it to someone unless they know they primarily want to live in Japan. I think it can be worth it, I feel like I changed quite a lot as a result of my 2 years there. The author's criticisms are valid but probably understated. I happened to work for a fairly wealthy lab that could pay me a salary, but this is not the norm for masters students, and even some PhD students don't get paid anything. Many, if not most graduate students I knew were working one or more part time jobs to make ends meet. I'm sure that the author's nice tech job and well-located home contributed to their enjoyment -- a lot of people have to live in Yokohama or west tokyo with > 1-2hrs of commute.
In my case, I had worked in Japan for a year at a domestic megacorp (which is how I met my prof), had attended a few conferences, was very fluent in Japanese, and knew with certainty that the enjoyment of living in Japan would make up for any objective drawbacks.
My main issue with grad school is that except for an exceedingly small number of labs that regularly aim for top conferences, the process feels very un-serious. I think the standards are low; very few professors care a lot about doing really good work; the classes are unnecessary, low quality (nobody cared about teaching much, in my dept), mostly surface level; and there's not much of a bar for student research. This might be exacerbated by the fact that the vast majority of undergrads begin research in their last year and continue into the MS program by default so there are generally a lot of less motivated students. I still had a few exceptionally smart colleagues, but I never collaborated with them (in fact, our lab has hardly ever published a paper with >2 authors). I generally felt that their hard work was misplaced.
About a year into my MS, after pushing a bit of slop at international conferences, I more or less gave up on working on my project and got a remote job based in the US. Like the article says, it is exceptionally convenient that you can work for 28 hours a week on a student visa, since 28 hours of a US tech job is more than enough to support a nice lifestyle for 1 in tokyo. I might genuinely have dropped out and moved back home if it wasn't for it. DO get the auth at immigration time, it's much more annoying later.
In any case, I really enjoyed it in hindsight. As a final note, Science Tokyo is very simple to enter for international students -- the only requirement is that you find a professor willing to take you in. There are many structures in place for international students and I'd say it's fairly nice and comfortable.
1. mcyc · 1790839330 · focus · HN ↗
  Ah, too bad we never crossed paths!
  Thanks for your comments here. I agree that a lot of labs were pretty unserious. One other thing that I should have mentioned in the article is that labs in Japan are like little contracting businesses, and the professors have to find funding in a way that seems much harder than the US. This leads to a lot of variability in the financial situations of labs, so yeah my concerns about funding/etc were probably understated (and my lab was likely one of the best funded in the nation).
  As for domestic conferences, typically my lab used them as staging for their papers that they would send to top international conferences, so the quality was still pretty good. I do get the sense that the quality outside of the T1 labs drops off pretty quickly.
  > ... but I never collaborated with them (in fact, our lab has hardly ever published a paper with >2 authors).
  This was something I had in the original draft of the article, but it was removed for being a bit niche. TiTech (and perhaps other schools, but not universal) had a restriction in place where only one student could claim primary authorship on a paper for the purposes of their graduation requirements (typically 1 conference + 1 journal paper). Even if two students contributed equally, they had to sign a form designating one of them the "true" primary author. This massively discourages collaboration. I absolutely hated this policy, but there was basically nothing I could do at an institutional level to change it. I ended up being able to collaborate pretty freely, since I reached my publication requirements quickly, but this policy really hurts students who do their PhD in Japan just because of the numbers game. If two candidates are up for a position, and one has N papers (all first author, but 2-3 authors max) and the other has N + 10 papers (with <N first author, but all with huge groups), the one with N papers is at a disadvantage at first glance. And it just removes one of the primary skills you need to cultivate during a PhD: collaborating with other researchers.
gumbojuice · 1790842915 · focus · HN ↗
I did a machine learning PhD while working in a smaller town in India. (not from US, but came from a research position at a lab at AAU in Denmark, my native place)
It's fun to read the piece, I worried about many of the same topics, but with completely different reasoning (e.g. I dislike megacities, crazy work culture, and I'm likely not google/stanford/top-tier material).
But in general I think going abroad for anything, not only a PhD, is a good exercise.
totetsu · 1790857816 · focus · HN ↗
I have a post graduate degree from a Japanese university. I thought the undergraduate program in my home country was far more academically rigorous. Most of my fellow students here seemed to be trying to just avoid having a hole in their CV while they waited to re-take licensing exams or find a job. I did get a lot out of the experience.. and I there is a lot of learning that can happen in that situation too, but I don't think anyones once noticed that I have this qualification.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.