LolWolf

16 ms·

151.
by LolWolf 6y ago
I think that's the joke of the title, though; perhaps rephrased slightly: "no matter how hard you try, this mathematical problem cannot actually be solved (and we can prove it, too)."
152.
by LolWolf 6y ago
It’s kind of the same, actually. You can treat the individual entries as “components” in a 2^d-vector with special multiplicative structure, which essentially transforms in specific ways. These axioms of GA lay that out more clearly: https
153.
by LolWolf 6y ago
> Why don’t research groups just publish to their own websites or directories like arxiv? Depending on the field, this is often the case. Here are some examples from both of the labs I'm in, with PDFs available: https://s
154.
by LolWolf 6y ago
I generally agree in the sense that "null results" should not be published as "results." But, especially in the experimental sciences, I think it would be an incredible (and very useful) feat of work to have well-documen
155.
by LolWolf 6y ago
> And now macOS is adopting the worst of iOS' draconian policies. You won't be able to run non-notarized apps on macOS on Apple Silicon based computers at all. Do you have a source for this? I certainly haven't read this a
156.
by LolWolf 6y ago
Ganondorf?
157.
by LolWolf 6y ago
The DOI link in the article is broken, but if anybody is interested, the original paper appeared in Science : https://science.sciencemag.org/content/369/6505/863 (paywalled) There are several websites that c
158.
by LolWolf 6y ago
Correct! I did not mean to say that it was actually Arecibo in a different country, but rather that it was fairly heavily inspired by it. I mentioned some of what you talk about in the latter part of your comment in another reply on a desce
159.
by LolWolf 6y ago
I don't think it is, since FAST (finished 2016) wasn't finished until well after the game had been released (2013)—on the other hand, parts of it were built (FAST was started 2011), so perhaps there was partial inspiration? I&#x27
160.
by LolWolf 6y ago
I mean, yes, obviously it's not literally the Arecibo telescope in the same location, but it's a heavy inspiration and quite similar in design and structure. (For example, an obvious difference in BF4 is the surrounding forest,
161.
by LolWolf 6y ago
It was also in Battlefield 4 ("Rogue Transmission")
162.
by LolWolf 6y ago
Same here, actually!
163.
by LolWolf 6y ago
A good way to make something readable to the general public would be to not even write it in math. (But then again, that's not a very useful statement when we need things written out in math.) Snarkiness aside, I do agree that a good
164.
by LolWolf 6y ago
The forwards and backwards in finder history shortcut also works in a large number of other apps :)
165.
by LolWolf 6y ago
It is also Command-Down to open the selected thing in Finder. Command-Up takes you one level in the file hierarchy (e.g., takes you to the parent), which also fits nicely.
166.
by LolWolf 6y ago
Oh, that would be awesome! I would really love to listen to a bit more detail about the choice of Julia for a company's stack. (I am an academic, so choosing Julia was very easy for me :) but I can imagine this choice could likely be h
167.
by LolWolf 6y ago
I think “this blog post” refers to the OP :)
168.
by LolWolf 6y ago
For sure! It’s a cool product (I got two people who messaged me about it this morning, before I even saw it on hn) and it’s pretty fun. Good luck with it! :)
169.
by LolWolf 6y ago
I think the embedding idea in specific is quite good and there are several potential approaches! One is, if you let xᵢ be the song vector for user i (whose kth entry is 1 if user i has song k in their library and 0 otherwise). Then you can
170.
by LolWolf 6y ago
That's fine; I don't think this is the general point. If you're trying to run a production system, need incredible robustness guarantees that somehow Julia cannot meet (which would be surprising to me), and require some speci
171.
by LolWolf 6y ago
This is what I meant when I stated > It's amazing to realize that writing down the first thing that comes to your head is usually like 80% as fast as a good, performant implementation. (As someone who has done a decent amount of wor
172.
by LolWolf 6y ago
Honestly, this is truly amazing. I will take a quick moment to thank the Julia developers (and all the contributors) for such an awesome language—honestly, if you do any kind of scientific computing (at all) I really and truly recommend try
173.
by LolWolf 6y ago
Oh, for a second I thought the GP was referring to nonlinear programming. Then I remembered this is an ML thread. There are some nonzero number of papers referring to NLP (as in nonlinear programming) in ML, mostly for the purposes of con
174.
by LolWolf 6y ago
Bounds for the best possible designs for optical devices: well-studied [0, 1, 2, 3, 4], yet really hard. More specifically, whenever you give a designer a design spec, it is always worth asking, how good is the best possible design for thi
175.
Training Convolutional ReLU Neural Networks in Polynomial Time(arxiv.org)
1 points by LolWolf 6y ago|0 comments
176.
by LolWolf 6y ago
Photonics and its impact in (classical and quantum) computing, VR/AR/Lidar, telecoms, particle accelerators, etc. There are some incredible advances being made in many areas of photonics, including in everything from how to perfor
177.
by LolWolf 6y ago
Author here! Let me know if there are any questions :)
178.
Sorting and information theory: why sorting is n log n(guille.site)
1 points by LolWolf 6y ago|1 comments
179.
by LolWolf 6y ago
> Yes, but we’re going to be restricted to O(L) final accuracy no matter what This is not, in general, true for smooth functions so long as L is small enough (you can reach arbitrary accuracy with GD if L is smaller than ~ the reciprocal
180.
by LolWolf 6y ago
> All optimal points Sorry I screwed up a bit when writing that: an optimal point exists which lies on the extremal points of the feasible domain. In many cases, it can be shown that the optimal point lies on the extremal points of t

More ›