currymj

27 ms·

361.
by currymj 6y ago
in computer science people can use pseudocode or descriptive variable and function names, and do sometimes, but still often fall back on math notation and Greek letters. sometimes the terseness, and leaving certain details implicit, does ac
362.
by currymj 6y ago
I agree with your analysis above -- a lot of the causal claims trying to connect <8hrs sleep to bad health outcomes are pretty dubious. People shouldn't worry about their lifespan or dementia or whatever. "Why We Sleep" un
363.
by currymj 6y ago
I need probably 7.5-8 hours of sleep. I'm very confident about this. If I get only 5-6 hours sleep for a couple nights, I become quite cognitively impaired; if it goes on long enough there are serious emotional effects too. There are p
364.
by currymj 6y ago
this raises the question, what are the companies where going into the office is like checking in to the Four Seasons?
365.
by currymj 6y ago
It's a common defense of the SAT that it gives a fair shot to smart students whose cultural background means they don't know how to play the admissions game. But I think this is not true at hyper-selective institutions, because as
366.
by currymj 6y ago
it's hard to show intellectual merit using the SAT, because the test isn't hard enough so people just max out the score. the 25th percentile student at MIT got 790/800 for the math section. it definitely does make a differenc
367.
by currymj 6y ago
The 25th percentile student at MIT gets a 790 math SAT, it's 730 for language. Difference between 790 and 800 is basically luck. It's at the point where the test can only hurt you not help you. There's no upside, but if you d
368.
by currymj 6y ago
a major part of 12-step programs is getting people to shift to an external locus of control. they're supposed to accept that they have a permanent problem that it's totally beyond their power to fix. somehow this helps people make
369.
by currymj 6y ago
Roam started doing something very close to this with its backlinks, and a lot of the competing notetaking apps are imitating this feature. For instance I believe Notion has this now too -- if you link to or mention a page, a backlink will a
370.
by currymj 6y ago
Isn't Notion pretty close to what's being described here? Sort of weird combination of networked spreadsheet, relational database, writing + notetaking app? There's been an explosion of competitors and imitators, too. Some, l
371.
by currymj 6y ago
do you work on high-performance computing, or more generally with numerical codes for scientific/engineering simulations? this seems to be the area where Julia is getting the most traction. it's strong in some other areas too. it&
372.
by currymj 6y ago
the book was appealing because it was something to point to that might convince your boss or friends to let you sleep, rather than acting like wanting to sleep more than 6 hours a night is a sign of weak moral character. many people hold th
373.
by currymj 6y ago
it's hard to say. there are indigenous Australian stories that have been preserved over 10,000 years that apparently have an accurate correspondence to the geological changes that took place over that time period. https://ww
374.
by currymj 6y ago
what kinds of deregulation would you hope to see? make it easier for small companies to do public offerings?
375.
by currymj 6y ago
yeah, people talk about "lottery-like stocks" in the academic literature on investor behavior. it's a standard term. i think the idea is that people buy lottery tickets or scratch cards for the same reason they do these stock
376.
by currymj 6y ago
I don't disagree with any of that, but I still think a responsible, clear-thinking ML practitioner can avoid having to assume the form of the data-generating process, depending on their application. In some cases if you care about PAC
377.
by currymj 6y ago
i think this actually gets at what makes applied ML distinct from statistics as a practice, even though there is a ton of overlap. statisticians make assumptions 1 and 2, and think of themselves as trying to find the "correct" par
378.
by currymj 6y ago
Fanny Holmes (wife of Oliver Wendell Holmes Jr.) once said, when asked why she belonged to the Unitarian Church, "In Boston one has to be something, and Unitarian is the least you can be." Unitarianism by the 20th century didn&#x2
379.
by currymj 6y ago
i see what you're saying and perhaps i shouldn't have been so negative in my original post. I think that people who want to write or code shouldn't be afraid to put imperfect work out there. But if to apply for a job you also
380.
by currymj 6y ago
your blog is very high quality and i've enjoyed it in the past (Metropolis-Hastings post especially). however there's a problem if everybody is going to do this, if it becomes standard that you want to have a blog in order to prom
381.
by currymj 6y ago
so the work being done is so important that people have to stay up all night to get it done in time -- it absolutely can't be postponed even by a few hours. but also, it must not be very important at all, because it's being left t
382.
by currymj 6y ago
almost all papers on arXiv are done in latex, you upload the source and they compile it on their servers. that’s a lot of papers! it’s most of math, CS, and physics, and some economics. it’s probably true that most papers aren’t written in
383.
EBF 483: Introduction to Electricity Markets(e-education.psu.edu)
2 points by currymj 6y ago|0 comments
384.
by currymj 6y ago
I use PyTorch for small-ish models, not really the typical enormous deep learning workload where you let it train for days. I still prefer to work on a GPU workstation because it's the difference between running an experiment in minute
385.
by currymj 6y ago
this just isn't true; if you skim some abstracts of papers in the field, you will see that in fact there is a lot of technical content. a lot of the big famous names in AI ethics (e.g. Moritz Hardt, Cynthia Dwork) also have very strong
386.
by currymj 6y ago
it's true, to some extent AI research is "whatever AI researchers do". whereas if work tackling similar problems is done by PL theory, operations research, or whoever, it's no longer AI. of course this goes both ways; du
387.
by currymj 6y ago
this just happens from time to time, it's part of mass psychology for whatever reason. people just have a moral intuition that short selling is bad. Read about Clarence Saunders and the Great Piggly-Wiggly Corner of 1923. Similar story
388.
by currymj 6y ago
AI Ethics isn't a non-technical field. See the accepted papers at the conference in question: https://facctconference.org/2021/acceptedpapers.html it's a mix of "soft" humanities, applied ML, method
389.
by currymj 6y ago
what university was this where there was an obligation to respond to letters from cranks?? how did that policy get put into place, that’s fascinating.
390.
by currymj 6y ago
the griddy model is that everyone should take a sort of second job as an electricity trader, but not make any money doing it. it wouldn’t be a problem if it were just a matter of paying a little more or less, that would be great. but there

More ›