aab0

35 ms·

301.
by aab0 11y ago
There's also the much-cited football problem: if high schools ended at a more reasonable hour like 3PM or 3:30PM, this causes problems for football coaches, who might see their after-school practices running to 6PM or something. I&#x27
302.
by aab0 11y ago
You could also ask what ended the kink. The timing looks like it matches up with the Thai floods, which killed price decreases for years afterwards. The experience curve means you need bulk to achieve improvements, and the floods shattered
303.
by aab0 11y ago
> And 9/77 of mainland Chinese are buying properties in Vancouver, so you multiply that by total number of people in mainland China (that are high net worth and affluent) and then multiply that by the average property value bought b
304.
by aab0 11y ago
"That's not even remotely what was done in the article by my reading." That's what it was described as in the grandparent: "This was a study based on "a Financial Times multiple choice survey of 77 high net wor
305.
by aab0 11y ago
This essay would benefit from making the particular anthropics used like the Self-Sampling Assumption explicit (and the paradoxical implications of it which leads to rejection by many) and explaining examples like the Doomsday Problem.
306.
by aab0 11y ago
Previously and better: https://news.ycombinator.com/item?id=11330587
307.
by aab0 11y ago
> Second, you would be surprised at the amount of information even small sample sizes can provide when there is great uncertainty about the true data distribution. This isn't about sampling error, this is about inverting conditional
308.
by aab0 11y ago
For me it was until the sports part. User chatter has nothing to do with future victories and the bookies would have priced in anything the public knows. Then it occurred to me that this sort of thing was hugely illegal and I noticed the fa
309.
by aab0 11y ago
> If you take the naive approach and assume it's just a matter of getting models with the same number of parameters as the brain has synapses, our biggest models are about a million times too small. And if you had that hardware, wha
310.
by aab0 11y ago
Youtube did something similar for 'interesting thumbnails' last year with deep nets (many uploaders do not specify a good thumbnail preview), and reported that it gave a nice performance boost.
311.
by aab0 11y ago
> Of course, it's hard to separate out or even infer the causes as nature vs. nurture (which the article is careful about) but the part about twins was the most striking for me: If you look, they already give you enough to get an id
312.
by aab0 11y ago
Seems consistent to me. Just because you have hardware doesn't mean you know how to use it well. Consider everyone accidentally writing an O(n^2) algorithm on their multi-gigahertz/gigabyte/terabyte (CPU/RAM/HDD) ma
313.
by aab0 11y ago
This has a much better title, though. When I saw that earlier link, I thought it was about some music library! While this tells me that it's animation software, in use by major commercial players, and not some obscure software no one c
314.
by aab0 11y ago
They also had several very specific pathways to the end result: if the calutrons didn't work for refinement, the thermal plant might; if the uranium didn't work, the plutonium might; if the implosion design didn't work, the g
315.
by aab0 11y ago
Summers: "Like experts in many fields who give policy advice, the authors show a preference for first-best, textbook approaches to the problems in their field, while leaving other messy objectives acknowledged but assigned to others. I
316.
by aab0 11y ago
Cowen's law: whatever it is, there's a literature on it. All those hundreds of thousands of research papers published each year have to be on something ...
317.
by aab0 11y ago
The money spent on CPU/GPU manufacturing improvement is a sunk cost and more than paid for itself along the way in economics benefits; the cost of the cluster that beat him is already around what he was paid for this one set of matches
318.
by aab0 11y ago
Which makes one wonder what exactly 'Lucid' is doing different from Cycorp Inc (1995, described in WP article for Cyc), and is exactly what the TR article doesn't cover. /sigh
319.
by aab0 11y ago
From the long run perspective, this is worse than the floating city. As they describe it, Sudan and Egypt only avoid regulating it as part of a legal trick. Once that problem resolves, the special status evaporates. At least when you are on
320.
by aab0 11y ago
"Reading this thread, I believe there's one aspect not discussed: in a battle between man and machine, it's debatable who wins and depends on the domain, but a man-machine combination always wins over both." It doesn&#x2
321.
by aab0 11y ago
As far as I know, using neural networks for function approximation destroys the various convergence guarantees available. NNs can easily diverge and have catastrophic forgetting, and this is one of the things that made them challenging to u
322.
by aab0 11y ago
That's an amazingly nice forum feature.
323.
by aab0 11y ago
> but the computer has had much more training samples as Lee Sedol could process in his lifetime. That's not obvious at all. I don't think you appreciate how rigorous and demanding the training of a Go world champion is, how ut
324.
by aab0 11y ago
It makes two references, actually: Cooper compiled 120 studies in 1989 and another 60 studies in 2006. This comprehensive analysis of multiple research studies found no evidence of academic benefit at the elementary level. It did, however,