bllguo

72 ms·

871.
by bllguo 9y ago
Could we not fixate on the choice of example? I'm not familiar with Belltown and I don't really care, I'm much more interested in the parent's point and whether it's true.
872.
by bllguo 9y ago
This strikes me as a utilitarian vs. deontological argument. The former: if everyone just played along and used things reasonably we would still be benefiting from cheap storage. The latter: stop playing games, "unlimited" is an u
873.
by bllguo 9y ago
I won't object to the fact that majorities can be wrong. There are many examples, ancient and recent. But that is a different argument from your other comment. I trust majority in science because it is rarely wrong. I'd need a b
874.
by bllguo 9y ago
consensus as in general agreement, as in a majority. Pointing out that some minority group exists is hardly a refutation. Would you not say there is a consensus the Earth is round?
875.
by bllguo 9y ago
How in the world is listening to climate scientists an "appeal to authority" fallacy? How are people supposed to get information about anything if this is your argument? Most of our knowledge comes from experts and specialists...
876.
by bllguo 9y ago
Yeah in general I find that cleaning the data, doing exploratory analysis, doing sensible feature engineering, etc. are much more involved tasks than running ML algorithm XYZ.
877.
by bllguo 9y ago
that is definitely a problem. But wouldn't backdoors only allow feds to catch cybercriminals after the act? Better security would prevent disasters in the first place. This seems to me a much better way of thinking about the issue
878.
by bllguo 9y ago
articles always overhype these types of advances. Specifically in materials science, they ignore: 1. ease of synthesis. Many of the cool materials with nice properties you read about in these articles cannot yet be produced easily at scale,
879.
by bllguo 9y ago
never a fan of people saying "X field is missing out on Y book" it's naive, and it's disrespectful to the field and its experts to claim that a book intentionally written to be accessible to laymen can invalidate the wor
880.
by bllguo 9y ago
Your last question intrigues me. I say "I." I'm curious whether other people use "we"
881.
by bllguo 9y ago
that's just how people operate. If that indicates HN is poisonous then all discussion places are poisonous, which really means nothing.
882.
by bllguo 9y ago
why would it be likely that daily active users ballooned 56% to surpass 2B? Seems like it must be monthly active in the discussion. Did FB do something recently that leads to your remark
883.
by bllguo 9y ago
Fun, will look into this! Wonder if anyone can point to other datasets?
884.
by bllguo 10y ago
exactly, that was my impression. Now I'm not so sure.
885.
by bllguo 10y ago
Embarrassing. This tripe gets published on The Economist these days? A piece on the fitness trend in SV without the trash psychoanalysis would have been somewhat interesting.
886.
by bllguo 10y ago
It's funny, I can see how being a doctor amplifies paranoia, since you now know of all kinds of horrible afflictions that could strike. I can also see how it reduces worrying since you are a trained medical professional.
887.
by bllguo 10y ago
baseless as of right now - they have yet to provide their evidence. They would be in a stronger position if they had released both accusation and proof together. The way they've gone about things is just fishy.
888.
by bllguo 10y ago
maybe he dislikes the "ahem, yours truly" part? Although I agree with your point, to me that phrase strongly conveys the "pseudo-proud" attitude the parent OP was referring to
889.
by bllguo 10y ago
refreshingly and surprisingly positive article about something that normally would be taken with much more negativity
890.
by bllguo 10y ago
To me it looks like she's defrauding investors and intentionally lying about Theranos' status and products. But sure, "soldiering on"...
891.
by bllguo 10y ago
> I thought all companies above a certain size were publicly traded. ...Wait, do I have some incorrect definition of publicly traded? Size doesn't really matter. I suppose you could say size and public status are correlated. Compani
892.
by bllguo 10y ago
Not believing that "a company of FB's size will implode in 10 years" doesn't imply belief in "a company of FB's size will never implode."
893.
by bllguo 10y ago
This is the only sensible reason I've read so far, from a comment in that thread: "There are behavioral biometrics solutions where profiles are built based on keystroke dynamics etc. which allow with a high degree of certainty to
894.
by bllguo 10y ago
I think things other than choice of algorithm are harder to copy. For instance, finding good outside sets sounds a lot easier than it is. But I do see your point. For me these competitions are more a fun educational tool where I can practic
895.
by bllguo 10y ago
Very interesting. Tangentially - it surprised me how ubiquitous gradient boosting was in these competitions, since the algorithm was never mentioned in my stats/ML coursework. Especially seeing how well they perform. I'm particula
896.
by bllguo 10y ago
So is it fair to say a large part of what separates the top placers in these competitions is feature engineering?
897.
by bllguo 10y ago
Wald's airplane damage story is a favorite example of one of my statistics professors. Humbling example of how statistics can easily be misinterpreted, at least for me. I can see myself making this kind of mistake (that the officers ma
898.
by bllguo 10y ago
Is this satire? If not you've ruined any point you're trying to make with that example.
899.
by bllguo 10y ago
Please, stop with the fearmongering. You are coming to some weird conclusions.
900.
by bllguo 10y ago
That's not really related to his point. Unless you are implying patenting a "patentable" discovery reflects how "scientists would rush to offer each of the technologies before they had adequately tested or even understoo

More ›