nonbel

15 ms·

91.
by nonbel 7y ago
I think AMD figured it out with chiplets actually... most software isn't even written yet to take advantage of cheap massive parallelism. That is an entire direction to grow after single thread performance flatlines.
92.
by nonbel 7y ago
AFAIK, Netflix isn't "profitable" either. As long as people keep giving them loans they are fine.
93.
by nonbel 7y ago
Stuff people like/need becoming cheaper to produce/attain is always a good thing. It means less waste of resources.
94.
by nonbel 7y ago
> "The race isn't to maximize profits, it's to survive the inevitable deflationary spiral in prices as competitors are forced to pass along cost savings to customers to retain market share." https://www.of
95.
by nonbel 7y ago
I thought the extra term was added so the equations didn't result in probabilities outside the 0-1 range. > "The Higgs boson gives mass to other elementary particles such as quarks and electrons, and its existence is necessary
96.
by nonbel 7y ago
Typo: ( p(H[0])p(D|H[0]) ) to the sum of all known explanations ( p(H[0:n])p(D|H[0:n]) ).
97.
by nonbel 7y ago
Yes, that is called a "degenerating research programme". The theory lags the observations.
98.
by nonbel 7y ago
Check out Imre Lakatos. "Falsifiability" is too strict since "every model is wrong". Lakatos calls that Popper0, which is not what Popper ever advocated. In a nutshell: The real thing going on is you are supposed to use
99.
by nonbel 7y ago
This is not my point here, but the galaxy you refer to is consistent with MOND: https://tritonstation.wordpress.com/2018/04/04/the-dwarf-gal...
100.
by nonbel 7y ago
I never said it was contradictory. I think what they've done is set up a system to filter out all theories except those that are flexible enough to be consistent with all possible observations. Specifically, they measure some value for
101.
by nonbel 7y ago
I am serious. This is what I have seen: Galactic rotation curves exactly as predicted: explained by dark matter Galactic rotation curves not as predicted: explained by dark matter Area of sky brighter than predicted: explained by d
102.
by nonbel 7y ago
>"The galactic center shines too brightly, like the glow of a metropolis at night where maps show only a town. [...] The new study, along with two others that came out in March, reopens the possibility that space-based instruments h
103.
by nonbel 7y ago
https://statmodeling.stat.columbia.edu/2019/03/20/retire-sta...
104.
by nonbel 7y ago
This is the best description of the main problem (testing your own vs some default hypothesis) I have seen: Paul E. Meehl, "Theory-Testing in Psychology and Physics: A Methodological Paradox," Philosophy of Science 34, no. 2 (Jun.
105.
by nonbel 7y ago
> You'd just calculate all the pairwise correlations and get nothing if they were solely totally nonlinear/interactions. I don't believe this. Most nonlinear correlations also show up as non-zero (linear) correlation coeff
106.
by nonbel 7y ago
I looked but did not see it.
107.
by nonbel 7y ago
>"The phenomenon that even random data will produce clusters." You don't really mean "random", you mean i.i.d. You can have a statistical model where the probability of something happens is random, but not indepe
108.
by nonbel 7y ago
Thanks, How about this "fact": The fact that these variables are all typically linear or additive ?
109.
by nonbel 7y ago
That is just a special case of the "everything is correlated" principle.
110.
by nonbel 7y ago
A model of whatever process you think generated the data. EDIT: I guess I should say that the concept of testing a "null model" without interpreting the fit relative to other models is wrong to begin with. You need to use Bayes&#x
111.
by nonbel 7y ago
> "It is true that, as Fisher points out, with enough samples you are almost guaranteed to reject the null hypothesis. " Where does Fisher point this out? > "That's why we tell students to consider both p values (w
112.
by nonbel 7y ago
That still gives 50-50 odds with sufficient sample size, not much of a test of the research hypothesis (since many alternatives will predict the same direction). It is better than 100% chance of rejection though.
113.
by nonbel 7y ago
This is mostly right, but you are still thinking of these rejections as "false positives" for some reason. They are real deviations from the null hypothesis ("true positives"). The problem is the user didn't test th
114.
by nonbel 7y ago
I had a post about how generic the signs were in that link and it got modded to oblivion and "flagged" so I guess no one had to be made uncomfortable by constructive criticism. What a great little microcosm of the larger issues wi
115.
by nonbel 7y ago
>" The fact that these variables are all typically linear or additive further implies that interactions between variables will be typically rare or small or both (implying that most such hits will be false positives, as interaction
116.
by nonbel 7y ago
As I said, no thanks to the signs. It looks like the NYT may have even purposefully left them out of their coverage due to how cringeworthy they are.
117.
by nonbel 7y ago
Your link says the "protests" were triggered by the NYT. It is not surprising they would cover the continuation of their own story: > "The backlash was prompted by an article in The New York Times last week" And I don
118.
by nonbel 7y ago
Someone on quora said they recently began trying to look again: >"Sir Robert Marx and I are now in the process of reinitiating the research we began in 1983 on this location where we found 3rd century Roman amphora. A sub-bottom pro
119.
by nonbel 7y ago
Whose idea is that?
120.
by nonbel 7y ago
Perhaps. Weren't they already known at that time though? >"Britain was known to the Classical world; the Greeks, Phoenicians and Carthaginians traded for Cornish tin in the 4th century BC." https://en.wikipedia.

More ›