pb1729

8 ms·

1.
by pb1729 5mo ago
Yeah, this is a good point. IIRC, I wasn't able to get the network to train very well at all with standard SGD. I don't think I thought to try Adam with β1 = 0, I will try it (& recompute brackets) if I get some time. If we ha
2.
by pb1729 5mo ago
If I understand your question correctly, the answer is that not only are the Lie brackets non-commutative, they're anti commutative (swapping the order negates the bracket). But this ironically means they end up having the same RMS, be
3.
The Training Example Lie Bracket(pbement.com)
33 points by pb1729 5mo ago|17 comments
4.
Three subtle examples of data leakage(lesswrong.com)
5 points by pb1729 2y ago|0 comments
5.
by pb1729 2y ago
It wouldn't cover steganography. What looks like an unencrypted video file may have an encrypted message hidden in the noise.
6.
by pb1729 2y ago
Bhauth wrote an analysis of this idea, and tldr is that trying to get an energy payback on literally vaporizing such a long cylinder of rock is brutally difficult, probably enough to make the economics of the plan unworkable. https:/&
7.
Fun with CellxGene(sarahconstantin.substack.com)
2 points by pb1729 2y ago|0 comments
8.
by pb1729 2y ago
tldr is that it happened because the universe cooled down from a stupendously insanely high temperature to a merely insanely high temperature shortly after the big bang. First look at this picture [0]: https://en.wikipedia.org&#x
9.
Hardware hedging against scaling regime shifts(old.reddit.com)
1 points by pb1729 2y ago|0 comments
10.
Well, I guess I believe everything now(blog.plover.com)
3 points by pb1729 2y ago|5 comments
11.
A Rigorous Derivation of the Bubble Sort Curve(linesthatconnect.github.io)
41 points by pb1729 2y ago|2 comments