bbctol

54 ms·

451.
by bbctol 10y ago
The underlying math was figured out a long time ago, but it's only been in recent years that we've had the computing power to test these out on lots of complicated, real-world classification problems, and had some incredible succe
452.
by bbctol 10y ago
I find Pynchon much more entertaining to read than Joyce. If he's telling a joke, the reader is certainly in on it; it's pretty clear he takes himself only as seriously as he should.
453.
by bbctol 10y ago
The author seems to be pulling things out of context in bad faith. It is unlawful, though not a criminal offense, in Australia to publicly say things likely to offend people of a different race, based on their racial group, provided the sta
454.
by bbctol 10y ago
The 18C provision of the Australian Racial Discrimination Act he mentions was brought in in 1995, a little early for the "millennials" he wants to decry. Nor does someone walking out of a speech strike me as even tangentially link
455.
by bbctol 10y ago
I'm afraid this is slightly off-topic: does anyone know what Atlas Obscura has been doing differently for the past few months? They've been around since 2009, and were a pretty quiet corner of the internet, but a few months ago th
456.
by bbctol 10y ago
Well, the actual paper tested experienced academics, grad students, and novices on their ability to differentiate, so you can measure that hypothesis of yours.
457.
by bbctol 10y ago
No--one of the reasons it's not a more accessible path is because it pays so poorly. Science funding doesn't follow a normal supply and demand framework; money isn't sent to scientists by customers in exchange for the benef
458.
by bbctol 10y ago
It seems much more that this is due to the rise of predatory publishers, where editors are willfully lax on standards, as the goal is not to make money but sucker scientists out of open-access publishing fees. Many prestigious journals just
459.
by bbctol 10y ago
That video is discussing a slightly different problem than the recently discovered issue the article's talking about.
460.
by bbctol 10y ago
The low pay for scientists is very definitely related to it not being an accessible path for lower classes.
461.
by bbctol 10y ago
You may not need a PhD or tons of experience to learn Deep Learning, but what about the gap between that and getting a job?
462.
by bbctol 10y ago
I'd be much more hesitant to talk about "meaning" in this context. Are we working with abstract concepts? We're working with images and text, and creating connections between the two. It may be that this approach, ramped
463.
by bbctol 10y ago
Nobody knows. Anyone born between 1980 and 2005 is at risk of the label. It just means "those dang kids with their computer boxes."
464.
by bbctol 10y ago
Right now there's a lot of different silos of post-publication comments sections, each with their own benefits and flaws. Leaving comments on PubMed is probably the best way to get your concerns noticed, but PubPeer is popular for its
465.
by bbctol 10y ago
There are lots of reasons for higher failure rates in one discipline than another that have nothing to do with how "scientific" they are. Cancer biology is indisputably a science, and has a higher rate of non-replicable results th
466.
by bbctol 10y ago
I think deep, natural language processing will be unambiguous: if you create a machine that says "Yes, I am intelligent, thanks for asking" in a way indistinguishable from a human, it would be hard to disagree. On the other hand,
467.
by bbctol 10y ago
I'll take the government's advice over the advice of a profit-driven business nine times out of ten.
468.
by bbctol 10y ago
I'm a little surprised that there's anything left to find on the wreck! I guess I've really underestimated the difficulty of underwater archaeology--I would think an important site discovered in 1900 would have been scraped c
469.
by bbctol 10y ago
To "observe" can mean any sort of interaction between the particle and a larger, connected physical system. To observe something, you have to have a physical interaction with it (it is hit by a photon of light which then hits your
470.
by bbctol 10y ago
There is a huge, huge placebo effect when it comes to cognitive enhancement, and while I understand that this is intended as a series of anecdotes, it still seems that the author doesn't fully understand this. Most notably, for his att
471.
by bbctol 10y ago
This seems like it puts forth fundamental limits of learning, in general. And even then, the only real limit is that sometimes, things are ambiguous, and pattern-matching is hard.
472.
by bbctol 10y ago
I'd say string theory has had more dancing proponents than supersymmetry by a long shot. Supersymmetry can be tested--that's what the LHC was for--and as the evidence mounts against it, proponents are right to change their theorie
473.
by bbctol 10y ago
Well, there certainly seems to be no way for anything to move faster than light through spacetime. It's harder to rule out exotic matter or negative energy that could warp spacetime to (relatively) move an object faster than light.
474.
by bbctol 10y ago
Perhaps you saw a slightly different version of the article, which didn't get as much click-through and so was replaced with this one.
475.
by bbctol 10y ago
It seems he doesn't like anything in general. Good resource, but the hardest thing about getting through LPTHW is reading it.
476.
by bbctol 10y ago
Every time I read Scott Alexander, I think there's either a Youtube video out there that explains with the same level of accuracy at twice the entertainment, or an essay out there that's just as long but goes twice as deep.
477.
by bbctol 10y ago
In terms of number of nodes, complexity of the arrangement of nodes, and amount of computation performed by each node, even a small biological brain is incalculably more complicated than a top ANN.
478.
by bbctol 10y ago
Was DeepMind 'lost'? It was bought by Google, but it's stayed in the UK (even if it might have made more sense to be folded into Google Brain) and iirc the founders are very keen that it stay in London.
479.
by bbctol 10y ago
How do you know that bacteria, or cement blocks, or hydrogen nuclei don't care about their existence?
480.
by bbctol 10y ago
I'd say perception is more relevant to the experience of unique identity than memory; I only see out of my eyes, which makes me feel separate from those around me, even if we'd have lived through all the same experiences.

More ›