cbgb

To add a different but orthogonal perspective to this:

My higher-level math courses were pure mathematics courses, and we pretty much always used Springer textbooks, which were only a few hundred pages long and the size of a normal paperback (i.e., not the size of, say, CLRS). When we didn't use Springer textbooks, we used other textbooks similar in size and length (e.g., [0]). I found these textbooks to be completely manageable to read as a student, and they were the best textbook-related learning experiences of my undergraduate years.

[0] https://www.amazon.com/Introduction-Galois-Correspondence-Ma...

While that may be true, I think what this boils down to is a form of prosecutorial discretion, where the discretion is in the hands of the ruling party.

For the record, I just read about the use of eagles (not falcons, though they are kept at falconries in France) in combating drones in a daily newsletter I highly respect. Here is the article they cite: http://en.rfi.fr/wire/20170220-born-killers-french-army-groo...

I guess I don't understand how at least the anti-drone eagles can be considered "fake news."

From the abstract [0]: "Conclusions: Endogenous IAP’s protective effects in regard to the metabolic syndrome may be inhibited by PHE, a metabolite of aspartame, perhaps explaining the lack of expected weight loss and metabolic improvements associated with diet drinks."

Can someone explain to me what I'm missing here? What is the "expected weight loss," and what are the expected "metabolic improvements associated with diet drinks"? As far as I'm concerned, one should expect neither metabolic improvements nor weight loss simply from consuming diet soda.

[0] http://www.nrcresearchpress.com/doi/abs/10.1139/apnm-2016-03...

edit: forgot footnote link

There's a lot of prep/clean-up I just don't want to deal with. I've tried this method before, and it just didn't work out.

All the washing of things you have to use in intermediate steps really adds up. And unless you eat chicken breast for every non-breakfast meal, 10 cooked chicken breasts will get gross in the fridge pretty quickly, so you'll probably want to do something like 3 breasts at a time if you're only eating one meal with chicken breasts per day. This means doing meal prep twice in the week, as opposed to once. If you don't refrigerate them, then you have to wait for the breasts to thaw, which is time I don't want to spend every day, multiple times a day potentially.

I have a busy schedule, and every hour counts some weeks, so the time saved after buying canned chicken breast really adds up.

Most people I know get grossed out when I mention this, but one way I saved a bunch of time re: all the chicken I had to eat for my bulking diet was to just buy canned chicken breast in bulk [0].

It's not nearly as gross as you think, and it saves so much time!

[0] http://www.costco.com/Kirkland-Signature%E2%84%A2-Premium-Ch...

This is just a nit, but Andrej Karpathy was never a professor at Stanford; he received his PhD from Stanford and now works at OpenAI.

See, e.g., Quine's "Translation and Meaning,"[0] as well as the blog post I linked to further down the page. Not every philosophy program consists solely of analytic philosophy; continental philosophy writing is rife with passive voice.

[0] http://faculty.washington.edu/stevehar/quine.pdf

The idea that passive voice is used in philosophy doesn't seem to be a novel idea [0].

For what it's worth, I don't think undergraduate CS degrees develop students' writing abilities at all. There are only a few proof-based courses I can think of, and those courses (e.g., Theory of Computation, Algorithms) need not necessarily be taught in a mathematically rigorous manner (which is to say, proof writing in these courses may be minimal, depending on the instructor).

[0] http://wehaveneverbeenblogging.blogspot.com/2010/02/on-activ...

Edit: grammar!

Philosophy/(Pure) Math double major here.

I would argue that my Math degree made me a "better" writer than my Philosophy degree did. Writing quality in philosophy is generally much poorer than other disciplines; you pretty much have to use passive voice everywhere, which is the first thing your college writing lab will tell you not to do.

On the other hand, in my Math classes, I had to turn in 5-10 pages of proofs every week/every other day for problem set work. It was this work that taught me how to write clearly and parsimoniously, much more so than my Philosophy courses.

Of course, YMMV.

Coincidentally enough, this was a submitted puzzler to the NPR syndicated show "Car Talk," except there were 10 men in a line wearing black or white hats. The goal was to determine the optimal strategy for guessing hat colors.

In fact, that's what NVIDIA did at my alma mater, Grinnell College. I believe the intent was for courses like the OS course to be taught using CUDA (at least to some degree). I don't think that has panned out, but now a tiny liberal arts college has a ton of GPUs to use.

I can't remember where I heard this, but I think the whole listicle thing generating a profit so that they could produce actually good journalism was always the plan.

I mean, I sometimes hear from their foreign correspondents on NPR or podcasts I'm interested in. They do have some great journalists on staff.

This argument, to me, seems like a lot of inside baseball. I think people attuned to the political conversation vastly overestimate how much political coverage the average voter consumes.

Occam's Razor might suggest people are voting for Clinton and Trump due to name recognition and previous primary results.

That's smart money. It turns out he's already suspended his campaign [1].

[1] http://live.mprnews.org/Event/2016_election_breaking_news (just search for the word "suspend")

To be clear, the posted course is not a survey course in machine learning. It is instead a more practical course on using TensorFlow to build deep neural network architectures useful for certain tasks.

The link the OP posted is a (great) survey course dedicated to machine learning as a whole, which includes methods other than deep learning.

I only doubt this theory because we civilians are not privy to strategic bombing locations. It is entirely possible that the US air campaign (or the Russians', for that matter) recently took out targets key to the financial strength of ISIS.

Ha, I haven't had the time to listen to it yet, but I was hoping my opinion would change as well. Glad to hear it's changed at least one person's opinion.

I don't really see how much good can come from this season of the podcast if they've chosen such a politically polarized subject. Surely they know there's very little chance the program will be judged on its merits; instead, I fear people will use it as another way to wedge themselves against one another, making for more unfriendly/uncomfortable political discussions between friends and family.

There's already been much ballyhoo about his role as a "traitor" and a "coward," and it seems very few on the political right even valued his life enough to trade him for a few prisoners. Is this podcast meant to change their minds? Studies typically show that facts make people more entrenched in their opinions. What, then, should we expect of the subjective opinions of the target of their derision?

EDIT: Quotes around 'traitor', 'coward'.

FYI: The second imgur link doesn't actually show any pictures. The first one looks great, though.

One technical reason may be that C++ (I only looked at LightLDA) has a higher learning curve than Python, which the Tensorflow docs stress.

Though Tensorflow stayed at #1 for a while, it only garnered around 200 comments, which is very high for a ML topic on HN (in my anecdotal experience, ML topics are highly rated but under-commented.) I imagine the audience for this library to be very small compared to Tensorflow, which likely included more ML/Google FOSS enthusiasts than day-to-day practitioners. Looks like this library is firmly targeted toward the latter.

I would hardly categorize Daniel Craig's refresh of Pierce Brosnan's Bond interpretation as "simple replacement." I think each new Bond actor has great potential to bring their own unique artistic interpretation to the role and to make it work for them.

I think that what you may be experiencing is a heightened sense of concentration toward a passion of yours that you were passionate about before you smoked pot.

However, think about children who smoke pot from an early age and don't develop the kind of curiosity that I'm sure proceeded the pot smoking you're talking about. Did you first discover your passions while you were young, impressionable, and smoking pot? Or did that come afterward, after all the initial experimentation and learning about yourself had already occurred?

I think the following quote from a NYC high school teacher better explains what I'm trying to get at:

“I hate pot. I hate it even more than hard drugs. I’ve taught high school for 25 years and I hate what marijuana does to my students. It goes beyond missing homework assignments. My students become less curious when they start smoking pot. I’ve seen it time and time again. People say pot makes you more creative, but from what I’ve seen, it narrows my students’ minds until they only reference the world in relation to the drug. They’ll say things like: “I went to the beach and got so high,” or “I went to a concert and got so high.” They start choosing their friends based on the drug. I hate when people say that it’s just experimenting. Because from what I’ve seen, it’s when my students stop experimenting.”[0]

[0] http://www.humansofnewyork.com/post/129574836736/i-hate-pot-...

This is pretty much exactly the message of a talk I attended at CU Boulder. The Microsoft researcher spoke of how linear regression for the classification of whether a given medical patient will be re-admitted to the hospital offered much more valuable insight into the relationships different diseases had to the probability of re-admittance.

Moreover, not-so-obviously strange rules learned by more complex models were exposed as strange by the information-dense pictorial representation of linear regression.

Just for the record, from the first page: "Background material needed for an undergraduate course has been put in the appendix. For this reason, the appendix has homework problems."

The appendix covers Probability and Linear Algebra.

When I was in college, I was once not allowed to join a Complex Analysis course because there were 15 students in the class. College guidelines advise against more than 12 students in a writing-intensive course, and because this was an upper-level Mathematics course, the professor took that to heart (he thought 15 was too much, but some needed the course to graduate).

This is to say that proper abstract Mathematics courses should require the students to write many proofs. I'm sure that students in the aforementioned Complex Analysis course were writing 4 - 7 pages of mathematical prose for their weekly problem sets. This is a non-trivial amount of writing practice which is especially tuned toward accurately expressing the interplay of precisely-defined abstractions (which all documentation should strive for).

Even discrete mathematics courses (Combinatorics, Graph Theory) should eschew simple calculations of permutations/combinations and graph traversal algorithm steps in favor of writing proofs of more of the abstract concepts. In this way, students will be trained to write effectively about abstract concepts, which will prepare them for a career in programming as well.

Ah, OK, thanks. I do get a kick out of reading various Google white papers (GFS, Spanner, and the Multi-Paxos implementation for Chubby come to mind), so I'd definitely be interested in the prospect of a future white paper :).

Do you have a link to that white paper? A quick search on their research site doesn't really yield any results.

This may be true for mathematics in the analytical tradition (things like Topology, Measure Theory, Real/Complex Analysis), but Linear Algebra is far more important in, unsurprisingly, algebraic disciplines. These include Number Theory, Field Theory, and Mathematical Logic.

While studying mathematics in college, once I had finished my Real Analysis requirement, I jumped headfirst into the algebraic side of things and never found myself using any sort of calculus. Even in my Topology course, we focused much more on using techniques from Real Analysis than specifically calculus topics (the former just being a generalization of the latter).

Probability theory is somewhat deceptive in its classification, since much of it "feels" a lot like a discrete mathematics course; however, much of the concepts, like you say, are underpinned by measure-theoretic principles, which is heavily analytic. It makes sense that calculus would come in handy in a much deeper study of probability theory.

Similar to another comment, how can you evaluate the correctness of the computed value if you don't know how the computation is done?