VodkaHaze

6 ms·

1.
by VodkaHaze 10y ago
I don't think it's a male fantasy. I also don't think it's biological. Just look at how women and men are socialized from a young age. Example: up until recently, many young girls in the west grew up on Disney princesses
2.
by VodkaHaze 10y ago
Now, now, I doubt FF has 700! ~= 10^174 CVEs
3.
by VodkaHaze 10y ago
The next step up from those CPUs really is going to be dual xeon boxes, which is a huge jump. Intel tried to do enough market segmentation that the curve from a Pentium all the way up to dual CPU workstations was fairly continuous in price,
4.
by VodkaHaze 10y ago
I wonder when the fixed costs of upgrading will be less than the operating costs of inefficient 15 year old hardware
5.
by VodkaHaze 10y ago
Importantly, a very strong communicator of scientific work. His passion shined through in the videos he made, which any non-expert could understand.
6.
by VodkaHaze 10y ago
> The only question that remains is: Stupidity or malevolence? Stupidity. On your end. Let me address your claims one by one. > the end of Bretton Woods Ending Bretton Woods was voluntary. The gold standard leads to volatile inflation
7.
by VodkaHaze 10y ago
I mean, even the Google style guide says to avoid complicated template metaprogramming, unless necessary
8.
by VodkaHaze 10y ago
The best I found are CppCon youtube videos, really
9.
by VodkaHaze 10y ago
Just so you know, there are really powerful "classical statistics" classifier models! Like ordered or multinomial logit/probit which you can incorporate latent classes into, etc. Reference book for those is here[1] [1] http:
10.
by VodkaHaze 10y ago
Thanks!
11.
by VodkaHaze 10y ago
In multiplayer zero sum game, it still works to Nash. Intuitively, in a zero sum game, running CFR on one player will find him best responses for the strategies he's playing against. Doing this for multiple players finds corresponding
12.
by VodkaHaze 10y ago
CFR is based on the method of "regret matching", which is a policy update method. Roughly, it says every iteration, the probability you take an action is equal to the fraction of its cumulative counterfactual regret over total cum
13.
by VodkaHaze 10y ago
Not really, because CFR is not only taking regret as the loss function, but also the method of "regret matching". That is making the mixed strategy probabilities in the next iteration equal the cumulative counterfactual regret (wh
14.
by VodkaHaze 10y ago
I agree. If you are going to bash C#, call it by its proper name, Java++ That said its still a fine language. Not a fan of the dogmatic OO design concept in Java and C#, but it's perfectly good at what it does.
15.
by VodkaHaze 10y ago
So you cherish the memories, not the thing
16.
by VodkaHaze 10y ago
> fling raw pointers around with abandon. Jitters in seat
17.
by VodkaHaze 10y ago
Do you still need the costly intel compiler suite to run some C++ code on it? Honestly they make it pretty hard to hack with as a device. It could succeed as a device if they let devs easily create cool applications with it.
18.
by VodkaHaze 10y ago
Moreover, most hard problems are simply proven hard because they incorporate a hard problem. Eg. Many proofs that say "this problem is NP-Hard" just prove that "you would have to solve xyz NP-Hard problem to solve this"
19.
by VodkaHaze 10y ago
Can't PM you, but willing to get one going.
20.
by VodkaHaze 10y ago
Julia is another option. You can even call R and python code from a Julia REPL. You generally want an interactive language, though, because there is an iterative cycle in prototyping models.
21.
by VodkaHaze 10y ago
See I'm fine with all of what you just said. The problem comes at the next step, "Testable claims are dubious so XYZ." Being skeptical is healthy. Inferring from your skepticism is not. Inferring leads to refusing vaccines,
22.
by VodkaHaze 10y ago
You can say that objective reality is dubious. But if you use that to subvert claims that survive testing, you're making a positive statement and you need a testable alternative. In any other case, what you're doing is using philo
23.
by VodkaHaze 10y ago
There's a difference between the positive (this thing is like that) and the normative (this should be like that). Positive questions are subject to empiricism, period. There is no cop out saying a positive statement is not testable. Ev
24.
by VodkaHaze 10y ago
The problem is that postmodernism tends to make positive claims, but there's no empiricism to back it up. If you're going to say everything is up to interpretation, you're making a positive claim. That requires empirical proo
25.
by VodkaHaze 10y ago
Get a team of high level economists in your central bank and economic advisory board. Start with curbing hyperinflation. That's been done before several times. Do what is known to work. There are some macroeconomic PhDs whose wet dream
26.
by VodkaHaze 10y ago
That sounds like a lovely workstation. I had a dream machine of overclocking a dual x5670 xeon machine, but motherboards that let you do that now are too rare and expensive.
27.
by VodkaHaze 10y ago
I mean for the average use case, a 5 year old pc generally has the compute capability. Gaming requires updating (though 32nm cpus are still perfectly fine). Scientific computing will move to the cloud soon enough, as you either need absurd
28.
by VodkaHaze 10y ago
The chip auto-overclocks, too, which seems really cool
29.
by VodkaHaze 10y ago
Disagree. Gate's main point is that the book is not rigorous, but it's based on a peer review literature (just making it accessible to laymen). So while some more technical people might get annoyed at the lack of rigor, it doesn&#
30.
by VodkaHaze 10y ago
Holy crap this is depressing. I'm happy that the state of the art in economics, currently, is to post working papers and preprints online. This has the side effects of papers being cited before being published, and turning out to be ba

More ›