rob_c

9 ms·

91.
by rob_c 11mo ago
and don't blindly believe a bbc headline these days...
92.
by rob_c 11mo ago
> because the AI's lack axiomatic systems beyond much manually constructed machinery. Oh god, that is massively under-selling their learning ability. These models are able to extract and reply with why jokes are funny without even k
93.
by rob_c 11mo ago
so your message is to short OpenAI before it implodes and gets absorbed into Cortana or equivalent ;)
94.
by rob_c 11mo ago
You're basically describing the knowledge problem vs model structure, how to even begin to design a system which self-updates/dynamically-learns vs being trained and deployed. Cracking that is a huge step, pure multi-modal trained
95.
by rob_c 11mo ago
most papers are also written in the same language, what's your point?
96.
by rob_c 11mo ago
tbf, transformers from more of a developmental perspective are hugely wasteful. they're long-range stable sure, but the whole training process requires so much power/data compared to even slightly simpler model designs I can see w
97.
by rob_c 11mo ago
so, exactly how much code/content is new/borrowed?
98.
by rob_c 11mo ago
The fact you're answering rather than ignoring shows the problem I'm very sorry to say
99.
by rob_c 11mo ago
yes, and they're totally not fighting on multiple fronts against central govt, those in the areas they're trying to run deciding that the civil service can now be a political arm of the left or from having to boot the actual racis
100.
by rob_c 11mo ago
tell me, did they install and set this up themselves or you?
101.
by rob_c 11mo ago
no, no, you can't that's just a factually vacuous and obviously incorrect statement...
102.
by rob_c 1y ago
This, I hope, is close to multi-modal in lingual terms. There's potentially a lot to learn from examining where this works/fails :D
103.
by rob_c 1y ago
I didn't say he knows a little, he knows a _lot_ clearly. I just think he puts on very rose tinted glasses when looking to the future rather than seeing the problems hitting ML model design/implementation now. We had a great leap
104.
by rob_c 1y ago
4 years of going through the algebra of back-propagation with maths and physics undergrads, it's not that difficult :). The main challenge is combining it with stats and almost infinite dimensions of freedom which makes implementation
105.
by rob_c 1y ago
> The problem is — nobody commits code that fails tests. Hah, if that were true the industry would be a better place. Or a worse place. Or a slower place but exactly the same. I should build a test for that... I've worked on many pr
106.
by rob_c 1y ago
> The code itself no longer matters. Good luck explaining that when you get hacked out of oblivion. This is like saying the fine-print of contracts don't matter so I get "AI" to regurgitate them all for me as a lawyer. It&
107.
by rob_c 1y ago
How tf else did you honestly expect black-boxes to get built, by self-mangling machine code spit out by a sentient AI god? Karpathy is bullish on everything bleeding edge, and unfortunately it kinda shows when you know the material better t
108.
by rob_c 1y ago
Trust but verify. It's not hard. The corollary being. If you can't (through skill or effort) verify don't trust. If you break this pattern you deserve all the follies that become you as a "professional".
109.
by rob_c 1y ago
tomorrow never dies
110.
by rob_c 1y ago
be careful, did you fill in the correct form to be able to make that post?
111.
by rob_c 1y ago
no, ofcom don't need to be picking the fights they are, they're choosing to support the political arm under the claims of "hate speech" and "ungood bad think"
112.
by rob_c 1y ago
it's asif we need to reform the system and find a better way...
113.
by rob_c 1y ago
it's been mulled over and keeps getting brought up again and again, the Overton window has shifted from "go away and come back when you're serious to", "christ how would we comply with that?" https:/&#x2F
114.
by rob_c 1y ago
If only
115.
by rob_c 1y ago
Not likely
116.
by rob_c 1y ago
Thanks for lumping me in with the great unwashed.
117.
by rob_c 1y ago
Under the UK goodness knows in principle yes but they have a protected status in terms of id and tax history (I, wish I was joking) which makes it problematic enough to even discuss...
118.
by rob_c 1y ago
I thought this was going somewhere rather than aiming to be a dictionary with pictures or am I missing a key paragraph?
119.
by rob_c 1y ago
Yeah that brings new definition to pebkac...
120.
by rob_c 1y ago
If it's the removal of the proteins not the plaque I wonder if that's the cause of people chasing the plaque for years? Classic correlation causation problem if so... Time to go read

More ›