tysam_and

13 ms·

91.
by tysam_and 3y ago
DL researcher here, it's also hard I think because experimentally many of us have noted (also some research on this) that there's a critical early phase to learning that conditions the network a certain way, and adding and/or
92.
by tysam_and 3y ago
Is this the new 500 mile story (or the second-place replacement?) I don't know whether this is true or not, but either way: incredible. Love this, and love the company for actually sending someone on company time (!??!?!?!?) to check i
93.
by tysam_and 3y ago
I think I read this a while back, maybe even on hacker news lol, it's a classic. Just like the 500 mile email story on the front page, though that one is _far_ more popular, hahahaha. :'))))
94.
by tysam_and 3y ago
It's an RNN, there is no N^2 component over time. It only requires the previous state. (there's a discord, you should join it with further questions! I unfortunately am not as informed as I should be on this one, other than the fa
95.
by tysam_and 3y ago
RNN inference on a smaller edge controller (all history is cached in a single state point for each layer, so much less memory and computation requirements IIRC) :') Very mobile-device and battery-powered systems friendly. :')))) ;
96.
by tysam_and 3y ago
This is what RWKV ( https://github.com/BlinkDL/RWKV-LM ) was made for, and what it will be good at. Wow. Pretty darn cool! <3 :'))))
97.
by tysam_and 3y ago
_The_ classic CS story (I personally think! ;) <3 :')))) ;'D). I think it will be hard to beat this one. :') Ah, the good ol' days....Wait, no, we're still living in 'em now! Just go to the edge of a quiet,
98.
by tysam_and 3y ago
It's a good story Brent
99.
by tysam_and 3y ago
Apologies, thank you.
100.
by tysam_and 3y ago
And hence median vs minimum, median scales well, but minimum is more aggressively equitable -- after all, it's a nice 'soft scaling lock', which we currently seem to struggle with. If we can hit a nice logarithmic return on i
101.
by tysam_and 3y ago
There is an equation with an equilibrium: CEO's pay rate = (some multiplier) * median salaried worker's pay rate OR CEO's pay rate = (some multiplier) * minimum salaried worker's pay rate We've been conditioned to t
102.
by tysam_and 3y ago
I would be surprised if that wasn't a glaring point for them, but you do control as best as you are able to. This should be the second of two Phase III studies, which are pretty rigorous. Some drugs are just very hard to make a valid c
103.
by tysam_and 3y ago
If you're talking about pairing an SSRI with MDMA, then perhaps you would perhaps be well-served by adding some reticence to the confident assertions made. That pairing would be like a Clorox + Bleach pairing of the psychotropic world.
104.
by tysam_and 3y ago
51%+ is a very powerful number.
105.
by tysam_and 3y ago
This feels pretty aggressive, I think we're on a different wavelength here. That's not quite what I'm saying, but it seems best for us to leave it at that.
106.
by tysam_and 3y ago
That's a fair response. I'm generally okay (some modern connotations aside) with people using the going brr bit. The servers where H.He spends a lot of time in actively use the meme when mocking outside groups (w/ few neutral
107.
by tysam_and 3y ago
I'd like to create it in the future, but I can share a rough version of what I have. Basically, as long as you gain more information than noise by scaling your experiments down (i.e., the smaller models create similar directions in the
108.
by tysam_and 3y ago
I did address those earlier in terms of association, I'm not sure if I understand the aggressive tone of this response however, or of a few of the assumptions in it. A few other people have made similar points about this culture as wel
109.
by tysam_and 3y ago
That's a great point, I hadn't really thought about that a lot, and definitely had not made the hard connection. I do love how he continues to spin out new potential modalities for Deep Learning. While underrecognized due to the h
110.
by tysam_and 3y ago
H.He has some overlap with the EAI community which has more crab-bucket tendencies when picking topics and/or mocking other groups. Even the more positive sentiments seem to be somehow crab-bucketed a bit (and, bizarrely at least to me
111.
by tysam_and 3y ago
First -- thank you for the thanks! It means a lot. :) Secondly -- same. It's the way my brain works. In a particular test I was administered a while back, 'coding' (i.e. mapping information transiently to symbols) was by far
112.
by tysam_and 3y ago
...what does this mean?
113.
by tysam_and 3y ago
I'm not sure if I follow the fitness-driving argument. Without an informational sieve of some kind, isn't it just random entropy? In this case, sure, far from equilibrium has more information, but that just means the system has mo
114.
by tysam_and 3y ago
Here is the simplest way of explaining the KL divergence: The KL divergence yields a concrete value that tells you how many actual bits of space on disk you will waste if you try to use an encoding table from one ZIP file of data to encode
115.
by tysam_and 3y ago
To be a little blunt, this comment does not make that much sense. Kolmogorov complexity is based directly off of Shannon's information theory. It's the definition of the shortest possible length of a program that can output a part
116.
by tysam_and 3y ago
I went up through Calc III, and of course PHYS I and II w/ calc is a barebones requirement. Discrete, linear, stats, the basics there. I think it varies per college. Calc II was the very first class I took on the very first day of my f
117.
by tysam_and 3y ago
Oh gosh, yes. I only had to 'upgrade' one class to an effectively harder version to get a minor in math. :'||||
118.
by tysam_and 3y ago
Oh my gosh, thank you so much. <3 :)
119.
by tysam_and 3y ago
Yeah, they are great and some of the reason (up the causal chain) for some of the work I've done! Seems really fun! <3 :)))) Facebook's Segment Anything Model I think has a lot of potentially really fun usecases. Plaintext desc
120.
by tysam_and 3y ago
Same.

More ›