eddotman

Wow - great paper! Very readable / accessible. I'm working on some stuff for NLP in materials science academic literature, but we haven't tried anything beyond the usual word embedding -> supervised classifier approach. I'll have to give this a try!

Pretty compelling arguments.

I do think that the ML / CS / etc. community is actually more open than other academic fields, and so this is definitely the right subfield to start in. Putting open access preprints online is not common practice in all disciplines, although it really should be.

I wonder if it makes sense for Distill to also publish on fields outside of pure ML - e.g. as applied to specific problems in other domains. I work in materials informatics, and I suspect that research in such fields (ML + applied sciences) might benefit quite a bit from having key results 'distilled' in this format.

Good call. Decided to do the same.

SEEKING WORK - Remote or Boston, MA

Freelance data science work (applied machine learning and NLP). I'm a PhD student at MIT in materials informatics (i.e. applied ML), and I'd love to do some interesting data science on the side. I usually work in Python, but I'm also comfortable with web languages (e.g. MEAN) if it comes up.

Consulting Website: http://www.dihedral.io/

Personal Website: http://eddotman.github.io/

Contact: hello [at] dihedral [dot] io

If OCW is too advanced to start, try Khan Academy first. It's great for K-12 material.

I've hung out with Ahaan at school several times -- some people in the comments here are speculating that going to university this young might make it hard to adjust (socially), but honestly I assumed he was like ~20 (i.e. just like any other college-age dude). I had no clue he had such a neat backstory of how he got to MIT!

Another Gryphon! I think you might be the first I've seen on HN. A few of my friends did Nanoscience co-op and I've pretty much only heard good things (I did the non co-op stream though since I didn't want to take the extra year to finish).

That sounds strange. Pretty much all the top STEM PhD programs (MIT, Harvard, Stanford, etc.) offer a liveable stipend and full tuition reimbursement with medical insurance. e.g. MIT gives like $30k/yr, which is reasonable to live on in Boston, and we never pay a cent of tuition.

I've never taken economics, but is it possible that this is sort of a combinatorial reason? Like, adding the N+1th person implies coordinating with N other people, so that's a quadratic growth. I'm not really sure if that's realistic though...

1) Sometimes big life decisions are no-win scenarios: there isn't always a solution that leaves every party happy.

2) It's easy to fall into the trap of thinking that what you're doing is 'harder' than what others are doing. Everything is hard if you're pushing your limits.

Yeah, I guess he can say a lot based on replays and Elo. Still, I feel like many people would feel awkward saying that they are better than their peers - even if it's true. Well, maybe that's partly a cultural thing; I'm sure it varies in terms of how uncomfortable people are about that sort of thing.

The brevity and bluntness in his responses is pretty amusing. I'm surprised he so candidly rates his own abilities and the abilities of his peers - I feel like many people would dodge those questions in public interviews.

Pretty cool stuff, dude. I filled out the form (which, I assume, sent you an email). I'll be around Berkeley in January, so if you like chatting about science stuff, then that could be fun. You have really good initiative (especially normalized for age).

I skimmed through it, and most of the advice sounds pretty solid. However, I suspect that following all this advice at once is not humanly possible (or at the very least, leaves you little time for anything else in your life). I did like the part about the 'cool people' secret paper club.

As others have said, $60k sounds steep for this. I wonder if these data science programs will catch on. Having done my bachelors in 'nanoscience' - I feel that these data science courses are a little buzzwordy in a similar fashion. I'm somewhat skeptical that these data science programs will exceed the value of a graduate degree in stats, math, cs, etc.

Sounds really cool. Are there any technical articles/papers available for the public to read on what you're doing? Or are all the details (understandably) under wraps?

The head researcher here, Yi Cui, is a pretty baller dude. I met him at a PhD admit visitation weekend, and his group was the most popular by far. Some of the admitted students joked that competition to get into his group was harder than getting into Stanford in the first place.

I suspect that his research success is partly due to some snowball effect though - he has an extraordinary amount of postdocs working for him and access to a lot of money/equipment - which are both pretty great assets to have in engineering (and tend to grow as a function of present size).

Anyway, not to take away from the research he's doing. A lot of the stuff he puts out is very cool; I look forward to seeing more.

Love the creativity in this - reminds me of the book Ready Player One, except with less 80s references and much harder (well, in some sense) puzzles.

I have a nice, clean version that I wrote up in LaTeX, and another 'fancy' version that I did in Illustrator.

You're right, there are many people (including you) who benefit greatly from prizes. But I'm more concerned about the mindset of attendees (i.e. those that "play only to win"). It's also worth noting that prize-winners are definitely in the minority, numerically speaking. Nonetheless, good point and thanks for the feedback!

Hey, this looks really great! Very clean layout. Looks like you have a typo under "competitor analysis" though.