superpermutat0r

9 ms·

31.
by superpermutat0r 7y ago
I thought that actual fatality rate is always higher than measured in case of exponentially growing pandemics. Just imagine the case where it takes 20 days for people to die. Your calculations of fatality rate this early in the disease woul
32.
by superpermutat0r 7y ago
Maybe they are too expensive. Austria has huge taxes for self-employed workers, especially if they are payed like engineers.
33.
by superpermutat0r 7y ago
I think in one of the algorithm lectures by Eric Demaine he outlines the idea of making a dynamic programming algorithm for good finger positions for chord progressions. Although you have to carefully design your cost function (moving the f
34.
by superpermutat0r 7y ago
I live next to a very busy road (6-lane). When I moved in, the closets were filled with black stains. My brother-in-law cleaned it and said he never saw that kind of dirt. The closets are dirty again. As time passed I also never encountered
35.
by superpermutat0r 7y ago
The breathing technique is a no BS technique. You just need to get the air in. You have to keep inhaling until you start feeling the effects of raised blood ph levels (dizziness, passing out, tingles etc.). You can breathe through nose or m
36.
by superpermutat0r 7y ago
This is not true, the breathing technique is pretty simple. You just breathe very deep and just relax and breathe out (not fully, just as much as comfortable). You breathe with a rhythm, faster or slower, anyway you like. After breathing fo
37.
by superpermutat0r 7y ago
But that adaptation is not a meat eating specialization. Stomach acids are rarely the adaptation, it is the digestive enzymes or gut bacteria that are the main drivers. And humans have none specialized for raw meat. Cows can easily live and
38.
by superpermutat0r 7y ago
But glycogen is not fat in the adipose tissue. Glycogen goes into muscles and is a ready fuel, fuel ready for fast metabolic burn. The metabolism to turn carbohydrates and protein into fat is insanely inefficient compared to just storing fa
39.
by superpermutat0r 7y ago
How can the body, given 5000 kcal in fat not store any of that fat in the adipose tissue?
40.
by superpermutat0r 7y ago
The thing about NN is that increasing the weights does improve the performance. The standard way to get good performance and see if your architecture works is just get the network huge (wide). After you see it works you get it small. The &q
41.
by superpermutat0r 7y ago
I also like the fact that if I have lets say $100 in various currencies (10 different currencies) and try to pay for something valued $95 in some other currency, Revolut will not make the purchase but Transferwise will autoconvert between t
42.
by superpermutat0r 7y ago
I think multidisciplinary knowledge is unnecessary, especially now with deep learning where there is no feature vector design. Even the DeepMind discoveries in various fields always feature the same people that I doubt know deeply about pro
43.
by superpermutat0r 7y ago
I remember reading somewhere about programs being turned into generating functions and then asymptotic complexity follows from that. Not just asymptotic in the worst-case but also the best-case and average case. It did not require a formal
44.
by superpermutat0r 7y ago
Looks a little bit like Jinc upscaling and some heavy denoising. I know it's NNs but still.
45.
by superpermutat0r 7y ago
They use statistical methods to eliminate the confounding variables. But given that they do not understand the fallibility of doing that on extremely noisy data, the conclusions are telling us nothing. They cannot control anything and it&#x
46.
by superpermutat0r 7y ago
There must be some hormesis from smoking. Young human body can adapt and work around stress, be it from smoking or exercise or alcohol. It is definitely a huge problem to maintain these physiologically stressful habits into old age when bod
47.
by superpermutat0r 7y ago
10x engineers make other 10x engineers, most of the time. Just look at mathematicians and their mentors and people they mentored, famous names everywhere! It's not a coincidence.
48.
by superpermutat0r 7y ago
Still, the nature of the control group might have a smaller taming effect needing longer time to turn into tame.
49.
by superpermutat0r 7y ago
The experiment is not designed well. I would guess that the tameness comes from the non stimulating environment more than the process of selection. It is known that mother imprints a lot of hormonal/stress activities to children while
50.
by superpermutat0r 7y ago
I think it has a more tractable algorithm than darcs.
51.
by superpermutat0r 7y ago
Thank you very much. I've already read more than 50 pages and I love it. It's just incredible how much problems can be now solved by a faster algorithm. I had an idea of solving lotto design problems, this might be a way to do it
52.
by superpermutat0r 7y ago
Anyone knows which are the four Japanese people he mentiones expanding on his idea, then him finding out about it, generalizing it even more, writing code and playing around with it and then writing the final report? He was being really non
53.
by superpermutat0r 7y ago
I'm pretty sure domestic cats kill more birds than windmills.
54.
by superpermutat0r 7y ago
People living over 100 years old obviously have huge genetic factors that allow them to have unhealthy habits (or at least unhealthy for the normal person). I mean, there are people over 100 that smoke every day.
55.
by superpermutat0r 7y ago
That's very interesting given that Tesseract uses Leptonica. I'm not sure if they use it for dewarping but all of my little projects with Leptonica really worked well. Dewarping, binarizing, extracting individual elements etc. ht
56.
by superpermutat0r 7y ago
She continues to add weight to the bar to this day.
57.
by superpermutat0r 7y ago
https://www.youtube.com/watch?v=k6GHHzIByOk Pretty simple way to reverse the osteoporosis.
58.
by superpermutat0r 7y ago
I do not even know about RenTech. I was just pointing out that there are several tiny hacks one can do to significantly expand capabilities of linear regression. Just like you can do polynomial regression with some data modification, you ca
59.
by superpermutat0r 7y ago
I do not see where you explained it. Doing cost sensitive linear regression (which is pretty trivial to implement) allows you to approximate Markov models, any hierarchical model, and all sorts of other stuff (like minimizing different cost
60.
by superpermutat0r 7y ago
Linear regression can be the reduction form of many sophisticated methods. For example, HMM or CRF like inference can be easily done with linear regression. Hierarchical models with several layers of symbolic inference can be done with line

More ›