dcl

8 ms·

91.
by dcl 3y ago
It's a reasonable way to look at it given that's how pretty much all 'deployed' versions of LLM's work?
92.
by dcl 3y ago
Exactly. But also, not just ASIOF is in the training set, but presumably lots of discussion about it and all the interesting events in the book.
93.
by dcl 3y ago
Similarly, consider a series like A Song of Ice and Fire. A human reader is still consciously aware of (and waiting for) the answers to questions raised in the very first book. This is millions of tokens ago, and that's if our brains t
94.
by dcl 3y ago
I've only ever used SQL and relational databases. What are the use cases of Key-value stores? What's the canonical example of where they are a clearly the right solution?
95.
by dcl 3y ago
I like to send these articles to my team to put our small mistakes in to scale.
96.
by dcl 3y ago
I don't understand how so much money has been poured in to these companies? I get the why the techniques are suitable, but I just assumed who ever wants to do this kind of retrieval can probably implement a suitable Approx. NN library
97.
by dcl 4y ago
It's great you've found techniques that work for you and you are correct the science is fuzzy... From https://en.wikipedia.org/wiki/Carol_Dweck#Criticism > Timothy Bates, a psychology professor at the Univ
98.
by dcl 4y ago
Would it be remotely possible to make a law to prevent this 'voluntary' deification of international despots?
99.
by dcl 4y ago
2 years ago I started a blog about statistics/ML and how they are used in business, purely to improve my writing and communication skills. I don't advertise the blog, and it's very likely had 0 hits, but knowing it's &#x
100.
by dcl 4y ago
What's wrong with a simple reciprocal policy, if they don't allow foreign owned/controlled social media in their state, we don't allow theirs here.
101.
by dcl 4y ago
She probably got an awesome new role though.
102.
by dcl 4y ago
110% my experience working in data science. I found it incredibly stressful to discover and provide analyses (even experimental results) that wasn't expected, or contradicted prior beliefs. The findings were always very harshly scrutin
103.
by dcl 4y ago
Spending more money on compute also makes you and the team look more important, an d the problems being tackled, more challenging.
104.
by dcl 4y ago
Would love to know more about how you produced the embeddings!
105.
by dcl 4y ago
Learned a lot of these lessons when studying statistics as a non-coder trying to implement my own techniques and other peoples research.
106.
by dcl 4y ago
I'm not aware of one, but an Australian hedge fund, Bronte Capital led by John Hempton keep their own database of fraudsters and their associated with the goal of identifying dodgy businesses to short. He has discussed this on a number
107.
State of AI Report 2022(stateof.ai)
1 points by dcl 4y ago|1 comments
108.
by dcl 4y ago
Always an interesting read
109.
by dcl 4y ago
As an ML person I totally get the how NN works to accomplish this and it's very cool. What's really cool is how they get this to work in real time with little/no latency.
110.
by dcl 4y ago
Only a small matter of time now before we have Vec2TikTok (or similar) where you'll be served up videos specifically optimised with the kind of content that will keep you scrolling or watching more. People will end up watching news abo
111.
by dcl 4y ago
I've worked at a bank. A significant proportion of the the workforce had no idea how it works. I'd say <10% of the senior executives have the slightest clue. But it's not 'as they wish' as you say, depending on w
112.
by dcl 5y ago
According to the wiki article on multi-armed bandits, apparently the Allies contemplated this in WW2. https://en.wikipedia.org/wiki/Multi-armed_bandit#Empirical_m... > Originally considered by Allied scientists in W
113.
by dcl 5y ago
This paper discusses that exact issue: Insurance makes wealth grow faster Ole Peters, Alexander Adamou https://arxiv.org/abs/1507.04655
114.
by dcl 5y ago
Yep one of those cases, where an ensemble average is not at all relevant for describing the situation.
115.
by dcl 5y ago
Chris Dixon ( https://twitter.com/cdixon ) posts some particularly outrageous and as far as I'm concerned, disingenuous comments on this topic. I would be ashamed to be associated with him.
116.
by dcl 5y ago
Indeed, I believe/think TikTok is probably the most psychologically harmful social media app.
117.
by dcl 5y ago
Obviously none allow it. But it's silly not fear the logical conclusion of the CCP's actions.
118.
by dcl 5y ago
True. But I fear China more.
119.
by dcl 5y ago
Chilling, and I suspect that western governments will do very little to prevent this kind of extra-territorial control exerted by the CCP. In a few years I expect it will be very increasingly difficult to criticise China.
120.
by dcl 5y ago
I remember that blog from when I was studying penalty/regularization methods over a decade ago. Was a great source of interesting papers and ideas.

More ›