jmmcd

20 ms·

211.
by jmmcd 7y ago
I see Belgium as an option on the region-selection page at the front of the website.
212.
by jmmcd 7y ago
Yeah, it's easier to run, but nah, because diminishing returns.
213.
by jmmcd 7y ago
> the individual is seeking to gain status Yes, but the individual may not know they're doing that . (I guess this is too obvious to the researchers to bother saying.)
214.
by jmmcd 7y ago
But that's not compounding in the sense of compound interest.
215.
by jmmcd 7y ago
Several references to simple linear models and even improper linear models , so I will recommend this paper: Dawes, The Robust Beauty of Improper Linear Models . http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.
216.
by jmmcd 7y ago
Profit is not quite the point. The point is that in order for HN to be a useful site, so that person n+1 decides to join, all they need is that the first n people have something interesting to say. For a Facebook competitor to be useful to
217.
by jmmcd 7y ago
> Before science fiction became respectable—indeed, inescapable—any fan could still tell you that Robert Heinlein and Theodore Sturgeon, and before them Stanley Weinbaum, wrote circles around Arthur C. Clarke or Isaac Asimov. Asimov has
218.
by jmmcd 7y ago
This is a different issue. We already make some assumptions about the relationship between training data and future data, even when not worried about interpretability, and these assumptions are enough for this paper's argument too.
219.
by jmmcd 7y ago
Without some juicy details of customers being screwed over, or name-and-shaming evil cackling developers, this story is a bit mom-and-apple-pie.
220.
by jmmcd 7y ago
Think Python, Think Complexity, and Think Statistics are all really excellent. Downey is an artist of CS pedagogy.
221.
by jmmcd 7y ago
Typo: x222 -> x220 I got an X260 on Ebay for ~EUR220 to replace an MBP whose screen went bad. The screen and touchpad are no match for the MBP, but it sparks joy in a way the MBP never did.
222.
by jmmcd 7y ago
Yes, but the point of Unix is that sweeping leaves and sweeping snow are the same task, when viewed in the right way.
223.
by jmmcd 7y ago
> it almost seems as if these were chosen retroactively to make the curve look the way it does. That's what regression does! And maybe they should've used regression, indeed. Instead, this looks more like an "improper line
224.
by jmmcd 7y ago
The article goes on to use `__import__()` to one-line imports.
225.
by jmmcd 7y ago
Well, I certainly agree that many people use ML in this way. But to me it's just bad terminology. The thing that happens during ML -- the L part -- must be the same as what happens during statistical learning. Many (sophisticated) ML a
226.
by jmmcd 7y ago
> doesn't quite cut it I think the underlying assumption here is that some approaches are too simple to "cut it", ie to be called machine learning. And eg PCA is too simple, and so should be excluded, and perhaps categoris
227.
by jmmcd 7y ago
There probably isn't one! But as argued in the paper, our reasons for thinking so are not based on NFL.
228.
by jmmcd 7y ago
Good! This article is about the NFL for machine learning (Wolpert), and also mentions NFL for optimisation (Wolpert and Macready). My recent paper [1] is about NFL for optimisation and mentions NFL for ML. The overall message is the same: m
229.
by jmmcd 7y ago
That's my point.
230.
by jmmcd 7y ago
It may be the camera is no better, but photos look better when shown on a bigger screen.
231.
by jmmcd 7y ago
> You don’t need to wonder what that world would be like, you live in it. Nah, there's a big difference between interbreeding, leading to a single species, and two non-interbreeding species living side-by-side.
232.
by jmmcd 7y ago
> a major reason how different human populations around the world became so different looking in such a short time. Is it true? Any link for this?
233.
by jmmcd 7y ago
I don't have a good answer. But I wonder if there are some realistic situations where we would have a good guess at the mean, but no clue about the distribution.
234.
by jmmcd 7y ago
Not to mention it's Radiohead . Some of the cynicism in this thread is so mindless.
235.
by jmmcd 7y ago
A few million they are donating to Extinction Rebellion.
236.
by jmmcd 7y ago
No, i indexes the patient, not time. (T_0, y_0) is one patients entire time series.
237.
by jmmcd 7y ago
No, I'm using "weaker methods" in the technical sense, not "performs worse". I agree that many papers of that recipe type exist -- Sorensen and Weyland have skewered them effectively -- they are just froth, to be ig
238.
by jmmcd 7y ago
Hmmm, maybe we're reading different papers. Ok, I admit that a lot of papers compare only against other metaheuristics, but only in cases where it is already accepted that those other metaheuristics far out-perform weaker methods. Abou
239.
by jmmcd 7y ago
> GAs in general are explored in academia because they are a fad that's easily publishable. I would say 90% of GA papers are because of this. > Other than that this class of methods is in general very computationally expensive an
240.
by jmmcd 7y ago
Slightly off-topic, this article has one of the best expositions I've seen (the Arthur-Merlin stuff) of P, NP, NP-complete, witnesses, and the idea of "evidence" for what complexity class the problem is in.

More ›