Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kxyvr
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
44 ms
·
301.
▲
by
kxyvr
10y ago
Well, from my perspective, there are three different components to machine learning techniques like multilayer perceptrons. Basically, 1. What's the model? 2. How do we define the error between the model and the data? 3. What algor
302.
▲
by
kxyvr
10y ago
Actually, what's funny is that this is probably the most succinct explanation of an operator overloading approach to AD that I've heard. But, from what I can tell as well, yes. Most of the action is in var.cpp: https:/
303.
▲
by
kxyvr
10y ago
This paper isn't really a great introduction into the concept of AD. That said, one way to visualize it is to see how a basic AD implementation works. There are generally two different varieties, operator overloading and source trans
304.
▲
by
kxyvr
10y ago
Thanks for the reply. Perlmutter and Siskind's paper doesn't really talk about the efficiency of higher-order derivatives either and only has a very simple run time for a gradient calculation. I had a look at the code and I can&#
305.
▲
by
kxyvr
10y ago
I skimmed their paper and don't claim to understand everything, but I am curious about some of your comments. As far as higher-order derivatives, there's not really been an issue with obtaining these derivatives with traditional a
306.
▲
by
kxyvr
10y ago
As a follow up to this, yes, 15 tends to be the standard deviation. However, most, if not all tests have a cap to the highest score they can return, which is about 160. I believe this is the case for both Stanford-Binet as well as Weschsl
307.
▲
by
kxyvr
10y ago
I work as an applied mathematician and in my opinion trying to work though the historical context can sometimes hurt. Basically, there are two issues at play. One, everyone has a finite amount of time to learn things, so working through h
308.
▲
by
kxyvr
10y ago
I find zero to be an interesting concept in the applied math world because depending on how you want to visualize it, it leads to different numerical methods. For example, think of how we represent a zero vector on a computer. Basically,
309.
▲
by
kxyvr
10y ago
Well, I suppose it's a philosophical thing, but I don't subscribe to the discover point of view. However, to be clear, I don't think there's anything wrong with this viewpoint. For me, math is a set of rules that we kno
310.
▲
by
kxyvr
10y ago
As a disclaimer, I've a Ph.D. in mathematics and work professionally as a mathematician. That said, this sentiment is something that I'm very sensitive to. I've a problem with your phrasing because I feel it implies that to
311.
▲
by
kxyvr
10y ago
There's an auxiliary reason to use Octave/MATLAB that most people don't talk about. Vastly cheaper access to high quality linear system solvers. Writing high-performing, multicore factorization codes is hard and only a few
312.
▲
by
kxyvr
10y ago
Prior to ACA, people who were old or obese could still buy insurance. Prior to ACA, people like me who were cancer survivors could not, in any way shape or form, buy health insurance outside of government programs. If you are asking wheth
313.
▲
by
kxyvr
10y ago
The current insurance market is not a permanent thing. If you look at the ACA marketplace, the list of plans changes year by year and that's the same thing that happens on the private market. In that context, what does keep your insu
314.
▲
by
kxyvr
10y ago
I'm in a similar boat. I'm a cancer survivor, in my early thirties, and run a technology business. Without government intervention, I can not buy health insurance. Even if you want to remove the compassion for it, eliminating th
315.
▲
by
kxyvr
10y ago
Unless something changed or I'm mistaken, I believe that System76 are just rebranded Clevos ( http://www.clevo.com.tw/ ). In the U.S., Sager has the contract for distribution ( https://www.sagernotebook.com )
316.
▲
Show HN: Optizelle, a library for nonlinear optimization
(github.com)
2 points
by
kxyvr
10y ago
|
0 comments
317.
▲
by
kxyvr
10y ago
As opaque and oppressive as it sometimes can be, I highly recommend that people at least understand the kinds of contracts listed in the FAR (Federal Acquisition Regulation): https://www.acquisition.gov/?q=browse/far&#x
318.
▲
by
kxyvr
10y ago
Not really. As a rule of thumb, a person at a moderately lean company costs two times their salary. A person working for the government costs around three times. The rest of that money goes to overhead, benefits, and taxes. More specific
319.
▲
by
kxyvr
10y ago
I think it'd be fascinating to look at as well, but I think it'll be hard to find one. Or, at least a quick search turns of nothing. Part of the reason why it's hard is how certain Latino populations identify regionally. I
320.
▲
by
kxyvr
10y ago
I looked at moving an account to Schwab, but it looked like they cut off their international wires. Did you receive any internal message about this? I may have been mistaken, but I recall when I looked a little over a year ago they offere
321.
▲
by
kxyvr
10y ago
You're correct that GWB received 40% of the Latino vote in 2004, but that was somewhat of an anomaly. From: http://latinovotematters.org/stats/ 1980 Jimmy Carter, 56% Ronald Reagan, 35% +21 1984 Walter
322.
▲
by
kxyvr
10y ago
Well, and that's the trick. Originally, I shied away from VC because I link into LAPACK. Even if I hook into a precompiled library from a manufacturer, they use the same name mangling scheme as Fortran. In order to figure that out,
323.
▲
by
kxyvr
10y ago
No doubt. However, in the past, we physically couldn't do what we can do now. At the moment, this is a EULA problem. If Apple changed a single line, that'd solve a ton of problems for developers and they can still have their pa
324.
▲
by
kxyvr
10y ago
I haven't tried in awhile, but is the best option just to use VC directly at this point?
325.
▲
by
kxyvr
10y ago
I use Homebrew and CMake as well, but it certainly has not been as smooth as my CMake scripts contain a series of fixes for MacOS. I also like CPack, but there's been less consistency with MacOS. Bundles don't work for multilang
326.
▲
by
kxyvr
10y ago
Paradoxically, I find it easier to port applications to Windows than MacOS. Due to the efforts from the Cygwin and MSYS groups, just about every build script ports naturally to Windows. As long as I use MinGW, the resulting application is
327.
▲
by
kxyvr
10y ago
That's the point. I don't want to have to ssh into it. However, since I can't legally virtualize MacOS on my laptop I'm forced to in order to make sure that my software works fine on MacOS. That's why I'm re
328.
▲
by
kxyvr
10y ago
So I'm one of those developers that got forced into buying a Mac Mini in order to test my software on MacOS. On one hand, you're correct that Apple "won" because I was pushed into spending $500 on their hardware. On th
329.
▲
by
kxyvr
10y ago
Absolutely this. And, it can't be emphasized enough if you want to keep your expenses down. As a preference, I think that this guy being able to make van living work is awesome. That being said, it's not necessarily cheaper than
330.
▲
by
kxyvr
10y ago
Slightly off topic, but your comment reminded me of Tarski's paper The Semantic Conception of Truth and the Foundations of Semantics ( http://www.jfsowa.com/logic/tarski.htm ) Basically, it's actually kind of
More ›