eutectic

17 ms·

181.
by eutectic 5y ago
This seems like exactly the sort of thing that would be susceptible to spurious correlation.
182.
by eutectic 5y ago
Seems like an excuse for introducing even more free parameters.
183.
by eutectic 5y ago
You can use the relation f(x) = 1 + integral_0^1 f(t) dt, then just check that e^x is a solution or use taylor expansion.
184.
by eutectic 6y ago
Unclear if question is genuine, but I would say that each level is capable of performing computations (and/or acting in the world) in ways which are inaccessible to other layers. I know saying 'things are hard' is an obvious
185.
by eutectic 6y ago
If it has a fan then it you might have to worry about aerosolized water.
186.
by eutectic 6y ago
Synthetic (cannabidiol, better known as CBD) not (Synthetic cannabidiol), better known as CBD
187.
by eutectic 6y ago
Or give the example of birthdays; in your first (initial) year of life you are 0 years old.
188.
by eutectic 6y ago
What about an enzymatic synthesis? Or even partial biosynthesis.
189.
by eutectic 6y ago
With 32 bits of precision I don't see it being a big problem over a maximum of say 50 moves. Neural networks which are not overfit should be tolerant of a small amount of random (not crafted) noise.
190.
by eutectic 6y ago
I don't think the integer weights are neccessary for sparsity; they are just faster because they allow for low precision. Of course floats aren't strictly associative so you wouldn't get bitwise equivalence between the increm
191.
by eutectic 6y ago
RAD's Charles Bloom has a blog ( https://cbloomrants.blogspot.com ) which is excellent for anyone interested in high-performance programming. Especially compression.
192.
by eutectic 6y ago
That would be illegal.
193.
by eutectic 6y ago
I was thinking power consumption. Probably a mis-correction anyway.
194.
by eutectic 6y ago
Oops, *mega, obviously! But I did mean watts.
195.
by eutectic 6y ago
metawatts?
196.
by eutectic 6y ago
According to google ngrams the frequency of 'satisfied that' is 0.00026% in British English and 0.00012% in American English. So only a factor of 2 difference. Interestingly the peak of American usage was in the 1860s (~0.001%), f
197.
by eutectic 6y ago
I guess it's more common in British English (and maybe a bit old-fashioned?). Edit: google ngrams seems to agree.
198.
by eutectic 6y ago
Well, that's truly scary!
199.
by eutectic 6y ago
I'm kind of surprised there's 800GB of text in the world.
200.
by eutectic 6y ago
You shouldn't ever have a collision with 128 bits of entropy.
201.
by eutectic 6y ago
After some thought I think it works as-is if you just delete all edges between points which can't directly see each-other, as long as the visibility graphs are still connected. The best cuts could then sometimes be edge->edge instea
202.
by eutectic 6y ago
Well,it is a set cover problem so you can use ILP and mip is actually much faster than z3. I'm sure it's possible (though maybe not easy) to use more geometry. I don't see how to make a simple linear program though.
203.
by eutectic 6y ago
I came up with basically this same idea independently when writing my solver (pasted in my top-level comment). It mostly works well, but it isn't obvious how to extend it to capture non-convex geometry. E.g. cutting off 2 ears to creat
204.
by eutectic 6y ago
I was inspired to write a solver for this kind of puzzle (it's not pretty but it seems to work well): import numpy as np import z3 import matplotlib.pyplot as plt from sklearn.svm import LinearSVC from sklearn.clu
205.
by eutectic 6y ago
Aren't dual numbers only useful for forward-mode AD, and not the more useful reverse-mode?
206.
by eutectic 6y ago
Optimal depth for union-find is O(1); every node points directly to the root. It's a very different case than building a binary search tree.
207.
by eutectic 6y ago
I think there's a paper somewhere claiming that completely random linking is just as good as union-by-rank (in expectation).
208.
by eutectic 6y ago
I would say it's a case of large surface area to volume ratio, inability to repair damage, and lack of selective pressure (why protect yourself from sunlight when you spend your time inside an animal?). No great mystery.
209.
by eutectic 6y ago
You could fix the means and the ratio of standard deviations, then binary search on the sum of standard deviations.
210.
by eutectic 6y ago
Quite fast and accurate: from scipy.stats import beta import numpy as np def betamax(params0, params1, integration_points=5000): u0 = params0[0] / sum(params0) u1 = params1[0] / sum(params1)

More ›