solomatov

8 ms·

61.
by solomatov 2y ago
A bit off topic, but what would you consider as the best applications of agentic ai?
62.
by solomatov 2y ago
Is it possible for you to add color theme/icon theme/keymap only extensions, without any executable code? I think, it will improve the security situation a bit. I don't see why the mentioned kinds of extensions should have an
63.
by solomatov 2y ago
Halting problem input has finite size (i.e. it’s a Turing machine)
64.
by solomatov 2y ago
LLM can never solve a halting problem (because no one can using a Turing machine).
65.
by solomatov 2y ago
You could have a rich mathematical knowledge, while being not very good at proving theorems. Also, you might be good at proving competitive mathematics problems without having a rich mathematical knowledge. It's also possible to have r
66.
by solomatov 2y ago
Thanks, very interesting information. I was aware of IC rules, but have never heard about limitations on model weights.
67.
by solomatov 2y ago
What are these ai diffusion rules?
68.
Ask HN: Deep learning model license enforceability
1 points by solomatov 2y ago|0 comments
69.
by solomatov 2y ago
> but luckily now that it's MIT licensed it's available on OpenRouter Did it have a different license before? If so, why did they change it?
70.
by solomatov 2y ago
A smaller to a medium sized company. Due to several reasons: - Setting up payment with a third party provider isn't that simple, and their fees are far from zero. - Getting users. Popular queries in Google are full of existing results,
71.
by solomatov 2y ago
If you are small, the app store looks to me as the easiest solution for selling apps.
72.
by solomatov 2y ago
My main concern is security and privacy. App store apps are sandboxed but manually installed apps usually are not.
73.
by solomatov 2y ago
I would be much more likely to install this if it was published in the app store.
74.
by solomatov 2y ago
My guess is they sample closed lambda terms representing functions.
75.
by solomatov 3y ago
Could you give examples of such companies (I am really curious)?
76.
by solomatov 3y ago
Drag (air friction) is proportional to the square of velocity. If you know aerodynamic properties of iphone, you could find a value of velocity where drag is equal to gravitational force, at which point the speed doesn't change. You co
77.
by solomatov 3y ago
We won’t need semantic web if machines are able to understand natural language.
78.
by solomatov 3y ago
Yep, you seem to be right. Google stores the quotes from pages, and it's fair use. Again, I am not a lawyer, and didn't think about this.
79.
by solomatov 3y ago
It's not a fair use question, it's a question about models reproducing the article text almost verbatim. (IANAL)
80.
by solomatov 3y ago
> state-space models make transformer based models obsolete We will see whether they work on a large scale pretty soon. I hope they will, but they might not be. There're models which might outperform more advanced models on the smal
81.
by solomatov 3y ago
Yes, it's doable. Your model won't be as large and as performant as a real large scale model, but you could train something. You could watch this for a start: https://www.youtube.com/watch?v=kCc8FmEb1nY&list=PL
82.
by solomatov 3y ago
It's much more complicated than that. The article 3 ( https://gdpr-info.eu/art-3-gdpr/ ) says two possible ways to get into a territorial scope of GDPR: - the offering of goods or services - the monitoring of behavi
83.
by solomatov 3y ago
>I am, personally... at least until it's all done on-device and functions offline. You could do it now. Apple computers with a lot of RAM are pretty good at running Llama2.
84.
by solomatov 3y ago
I am not a lawyer, but in US the federal law consists of 2 parts: US Code (USC), and Code of Federal Regulation (CFR). The USC is created and maintained by the Congress. USC delegates to federal agencies implementation of USC through CFR, i
85.
by solomatov 3y ago
You could upload image in the visual variant of the model and ask to describe it. It’s different from Dall-E 3
86.
by solomatov 3y ago
I tried it, and it's super awesome. Try taking an image, ask to describe it, and feed the description back. I have never seen anything similar.
87.
by solomatov 3y ago
Of course. You don't need to investigate the license, learn about the vendor, inquire about payment structure, talk to sales people, justify the purchase to your manager, etc. There's much less friction, all thing being equal.
88.
by solomatov 3y ago
It’s more or less the same thing. You could consider transformer a GNN.
89.
by solomatov 3y ago
> True...it's just difficult to draft a license that says something along the lines of "If you're Big Tech, we have the right to go after you, if you're small startup, you're safe". Take a look at the licens
90.
by solomatov 3y ago
There's a case law which contradicts what you say: - https://casetext.com/case/toyota-motor-sales-v-tabari (about using buy-a-lexus.com and buyorleaselexus.com domains without permission from Toyota) - https:&#x2

More ›