MrUssek

6 ms·

1.
by MrUssek 2y ago
I suppose Ciechanowski is the kind of blogger who causes an HN cascade with each new article.
2.
by MrUssek 3y ago
Some context for those that may be unaware; Chet has been a staple in the Android developer community for a very long time, having hosted a popular podcast and been a staple a conferences. His move into comedy will surprise no one, but he w
3.
by MrUssek 3y ago
While they are out of date and not too technical (and arguably, quite biased), the two books that I've enjoyed the most on this topic are: 1. Dark Pools, by Scott Patterson, and 2. Flash Boys, by Michael Lewis.
4.
by MrUssek 3y ago
Just to confirm my understanding; would you consider at least part of your product offering to be similar to temporal.io [1]? Your examples are reminiscent of theirs. [1] https://temporal.io/
5.
by MrUssek 4y ago
I use Feedly, not sure why it's not more frequently recommended. https://feedly.com/
6.
by MrUssek 4y ago
To be clear this is a forward mode auto diff implementation, not reverse mode, as might be inferred by the reference to the SPJ talk, correct?
7.
by MrUssek 4y ago
This post feels identical to one that I had been considering writing for a while now. I'm nearly 28, and the past ~4 months have felt like I have been suffering from acute brain fog. And not even regarding challenging topics; most of t
8.
by MrUssek 5y ago
Plus one for these. I found them a quicker and perhaps funner read than CLRS which was my undergrad textbook, and it covers a lot of interesting stuff. These are all a part of David Kopek's book project you can find here: https:/
9.
China has trained a 10 trillion parameter language model(twitter.com)
4 points by MrUssek 5y ago|0 comments
10.
by MrUssek 5y ago
Typo: Binutlis => Binutils
11.
What is your backup if the tech industry crashes?
4 points by MrUssek 5y ago|10 comments
12.
The Future of Deep Learning Is Photonic(spectrum.ieee.org)
1 points by MrUssek 5y ago|0 comments
13.
Separating MNIST digits using Optimal Transport(mrussek.com)
1 points by MrUssek 5y ago|0 comments
14.
by MrUssek 5y ago
There should be an option for StyleGAN (i.e https://thispersondoesnotexist.com/ )
15.
Enigma: GPT-2 trained on 10K Nature Papers: Can you spot the difference?(stefanzukin.com)
183 points by MrUssek 5y ago|105 comments
16.
by MrUssek 5y ago
Ok, I know it's crazy, but I like to indulge myself sometimes: Does this have something to do with Will Hurd's past in the CIA? Feels like an interesting choice to have on the board of the premiere AI research lab.
17.
by MrUssek 6y ago
So, uhh, where's the paper? The link in the readme isn't active.
18.
by MrUssek 6y ago
Assuming this is a hoax, it has to be one of the best timed hoaxes in history to be honest.
19.
by MrUssek 6y ago
I think that Errant Signal's video on Civilization offers a really interesting counterpoint to this view. When you think about it, despite it's cheap cosmopolitanism, it has an extremely narrow contemporary liberal view of civiliz
20.
by MrUssek 6y ago
This reminds me of a POC I did to serialize Kotlin continuation state in order to achieve something similar with Kotlin coroutines. The main issue with that approach is that the continuation context can end up closing over a huge variety of
21.
by MrUssek 6y ago
There's an extremely large number of parameters per "neuron". The 600B parameters will take up more than 1TB of space in memory, far too much for the 2080 TI or even main memory for most systems.
22.
GShard: Scaling giant models with conditional computation and automatic sharding(arxiv.org)
112 points by MrUssek 6y ago|35 comments