make3

20 ms·

841.
by make3 5y ago
the argument was about wether humans and neural nets learn in a similar way. I don't see how what you are saying has any impact on that
842.
by make3 5y ago
This intuition is very dangerous and leads to huge misconceptions about deep neural nets. Neural nets don't learn anything like us, and they don't reproduce our functions. We build on massive amounts of general symbolic knowledge,
843.
by make3 5y ago
Why is the one only example of the font available in the README in an agressive diagonal? Just let me read the damn thing! This seems a bit ridiculous to me.
844.
by make3 5y ago
Not sure why you're getting downvoted. Energy is a money game.
845.
by make3 5y ago
The comments here are atrocious.
846.
by make3 5y ago
wtf man
847.
by make3 5y ago
compound interest is interest proportional to how much you already have. This is like fixed absolute interest. Not sure why I was downvoted.
848.
by make3 5y ago
(this is not compound interest.. )
849.
by make3 5y ago
I'm not an American, saying something else is bad doesn't excuse the first one
850.
by make3 5y ago
make something up
851.
by make3 5y ago
Why do you say that the model is overfitted? You have no way of knowing that. Plus, 84 743 is a very reasonable size for a vision dataset with a binary prediction
852.
by make3 5y ago
like you did just now. That person made something up. Let's not put all Ph.Ds in a bag just for that.
853.
by make3 5y ago
future work is almost always discussed in publications
854.
by make3 5y ago
This country is a pure, violent dictatorship that imprisons more journalists and political opponents than anywhere else on earth, that is effecting ethnical k cleansing within its borders and that is aggressively looking to expend itself (T
855.
by make3 5y ago
Even with GPT-3, where this "game" would be much harder, this would be kind of a weak demo, because the human reader doesn't understand what the text means in either case (most of the time), taking much if not all away the in
856.
by make3 5y ago
I'm surprised by how broken the english of GPT-2 is. A lot of sentences are just broken. I would be curious to try again with GPT-3.
857.
by make3 5y ago
for most people what you describe is even more expensive and impractical than buying a mined diamond
858.
by make3 5y ago
I don't think it's that clear. The interface of a library is still part of the library, albeit a very small one (smallest possible to enable interoperability). So, the SCOTUS accepts that re-using part of a creation can be fair-us
859.
by make3 5y ago
The title is not clickbait, the reason the Warhol Foundation is using is what I think the parent is arguing is "clickbait".
860.
by make3 5y ago
how would that work, once js has information it can send it anywhere
861.
by make3 5y ago
patents give motivation to companies to invest in research. without patents you possibly would have neither of those things.
862.
by make3 6y ago
Yes, have you tried Julia? It is pretty much how you describe. There are other things that I don't like about that language, but the matrix part is really great
863.
by make3 6y ago
One of the first things you learn in theoretical sampling classes is that polls where people self select to participate (like twitter polls) are garbage
864.
by make3 6y ago
also they don't adopt them, they fund their caretaking. this title is pretty poor.
865.
by make3 6y ago
We should stop calling deep learning models AI, it gives unnecessary hype and misunderstanding of the capacities of the things, as well as a feeling that researchers aren't being honest about the capacities of the things they are devel
866.
by make3 6y ago
The actual (socialist) American left (think AOC, Bernie Sanders, Elizabeth Warren) calls this liberalism, and complains that it is more about posturing than actual change. So this is only a criticism of the left if you are very far to the r
867.
by make3 6y ago
plug a usb thermometer and write a script to throttle the hash rate when the temperature is high
868.
by make3 6y ago
cached because the blog was hugged to death: http://webcache.googleusercontent.com/search?q=cache:Awfh5p8...
869.
by make3 6y ago
I'd be very curious to see a source about your claim that Python is slowing down for uses outside of deep learning, I'm pretty sure you just made that up and it's false, not to be blunt
870.
by make3 6y ago
there have been Julia bindings to Tensorflow for a long time, and you can just use the Python ones in Julia with PyCall anyways. People haven't been doing that.

More ›