chroem-

32 ms·

331.
by chroem- 7y ago
I think the biggest advancement that could come out of the ML space right now is the equivalent of Big O notation for ML. Our current algorithms suck, and the first step to making them suck less is to precisely measure just how much they s
332.
by chroem- 7y ago
There's no downside risk. The USG and Fed will do everything in their power to keep us in a continuous state of "numbers go up." Borrowed billions of dollars to buy your own stock, and now you're underwater? Don'
333.
by chroem- 7y ago
Please, please, please someone fund DRACO antivirals [1]. It is absolutely shameful that they haven't been developed further because of a lack of funding. DRACO would do for viruses what penicillin did for bacteria. [1] https:/
334.
DRACO (Antiviral Drug)(en.wikipedia.org)
2 points by chroem- 7y ago|1 comments
335.
by chroem- 7y ago
Most active traders don't beat the market (any market), as they shouldn't provided that the EMH is true. I'm proposing a strategy of long term investing.
336.
by chroem- 7y ago
That's why they're average annual returns, and the idea is to hold for multiple years . The same could be said about people who bought stocks and real estate in 2007.
337.
by chroem- 7y ago
>I'm not saying Bitcoin doesn't have speculative value, but you have to be already an experienced analyst and trader If you buy and hold for multiple years, on average you will make 350% a year, very similarly to how you would
338.
by chroem- 7y ago
Even taking into account the bear market of the last two years, bitcoin still has an average annual return of 350%. To say that all cryptocurrency equates to gambling is just foolish.
339.
by chroem- 7y ago
I suppose that explains why a lot (most?) of the people who died also suffered from high blood pressure.
340.
by chroem- 7y ago
I'm sorry, but that's just plain wrong. The first US case at Providence Everett got well in a few days thanks to antivirals.
341.
by chroem- 7y ago
It's one of those things you can't really appreciate until you've tried it. Imperative programming languages are to assembly languages, as array programming is to imperative programming: it's a higher level of abstracti
342.
by chroem- 7y ago
Instead of describing the control flow of a program, you describe a set of operations on data structures.
343.
by chroem- 7y ago
Correct me if I'm wrong, but this seems to be missing array programming (APL, Matlab, Julia, etc).
344.
by chroem- 7y ago
Everything is hard, but software is a lot more vocal about it because it facilitates the modes in which we communicate
345.
by chroem- 7y ago
It's usually a generic checkbox on the same order as the "We care about your privacy!" tracking notifications.
346.
by chroem- 7y ago
The more difficult it becomes to commute from cheaper outlying areas, the higher rents will rise within the city proper.
347.
by chroem- 7y ago
Does the chip still manage to have superior power efficiency versus x86 even at these performance levels?
348.
by chroem- 7y ago
> Summary: we don't understand why programmers are paid so well. Isn't it just Jevon's Paradox but applied to labor?
349.
by chroem- 7y ago
It's interesting how the language around these incidents has shifted to give the impression that cybercommandos have stormed into cyberspace with their cyber assault rifles, when in reality the chances are very high that some universit
350.
by chroem- 7y ago
> High speed rail could share track with commuter lines in some cases Having ridden those very commuter lines, I can say that is not an option here. They are limited to 40mph along those routes because they follow a very windy path alon
351.
by chroem- 7y ago
The Link light rail project won't even reach Everett until 2036. That's about 40 miles of track in the 40 years since the project was approved. How is this supposed to work?
352.
by chroem- 7y ago
Protip: an EWMA is a smoothing convolution whose deconvolution can be computed in closed form with a little algebra. To reduce the effects of noise, apply the EWMA, then use a forecasting method of your choice to predict the smoothed serie
353.
by chroem- 7y ago
Dating apps are extremely broken, and yet they're almost all owned by a single company: Match Group. I have a suspicion that these apps are left intentionally broken to keep users fruitlessly searching and boost engagement metrics. I
354.
by chroem- 7y ago
Ideally you want to tackle societal problems before they become intractable. See also: the mess that ad tracking analytics has grown into.
355.
by chroem- 7y ago
I'm sorry, but "opt-in" has almost always been a load of BS in my experience. Most examples I have seen look like: "Do you accept these terms? A.) Yes, and continue using our product B.) No, and revoke your access to
356.
by chroem- 7y ago
Yes, it's an extension of linear regression. You can incorporate different basis functions to model trend, seasonality, the effects of external regressors, different frequency components, etc. It gives you a lot more control over the
357.
by chroem- 7y ago
Basis function regression is a very under-appreciated method for producing time series forecasts. I've found that it beats most of the methods described in this article. Maybe I should make a blog post...
358.
by chroem- 7y ago
Who regulates the regulators?
359.
by chroem- 7y ago
As someone who likes to crack jokes at the expense of goldbugs, Dalio saying to buy precious metals got my attention.
360.
by chroem- 7y ago
Source? I would like to keep up on that.

More ›