cschmidt

10 ms·

91.
Large, creative AI models will transform lives and labour markets(economist.com)
1 points by cschmidt 3y ago|0 comments
92.
by cschmidt 3y ago
Another interesting article on the topic here: https://tedgioia.substack.com/p/a-scandal-is-taking-place-in...
93.
by cschmidt 4y ago
Finally replaced this week! https://www.underconsideration.com/brandnew/archives/new_log...
94.
by cschmidt 4y ago
The Economist has a whole series of this type of article, basically what to read to learn about X. https://www.economist.com/the-economist-reads
95.
by cschmidt 4y ago
You're right, it has changed since this morning.
96.
by cschmidt 4y ago
I'd like that as well. For a while I thought https://jazzed.com/ would eventually be this, even thought it was just available in the UK. But now they seem like they've moved to focusing on live performances.
97.
by cschmidt 4y ago
I hadn't looked at Primephonic in a while, but their site is now just a letter on how they're working with Apple: https://www.primephonic.com/
98.
by cschmidt 4y ago
I got my invite, and had a bit of a play. It seems like it hallucinates more than ChatGPT. There was something wrong with many answers. The prose it generates seems a bit more clunky and repetitive. It doesn't do any coding or prog
99.
by cschmidt 4y ago
It is their version of ChatGPT: https://blog.google/technology/ai/bard-google-ai-search-upda...
100.
Welcome to Codon, a high performance Python compiler(docs.exaloop.io)
2 points by cschmidt 4y ago|0 comments
101.
by cschmidt 4y ago
I’m a paying customer. They have been building up a system with new features over time that Faiss doesn’t support. At least from a customer view I’d be very surprised if they were using it. I may be wrong but I thought it was originally sta
102.
by cschmidt 4y ago
What are you saying is false? That they aren't using FAISS? Their whole startup is built around a proprietary ANN solution, which isn't using FAISS.
103.
by cschmidt 4y ago
https://serialpodcast.org/ from the This American Life (NPR) people
104.
by cschmidt 4y ago
I've always heard this is intentional with scammers. Their goal is to get you to call the number, so they have a cost if the user responds. Having some errors filters out the less gullible, who won't fall for the scam and will j
105.
by cschmidt 4y ago
It is a really wonderful interactive tutorial. So well done.
106.
by cschmidt 4y ago
So what is a better alternative for a retail investor to do top level portfolio optimization?
107.
by cschmidt 4y ago
No not all. They rebalance only to get you back to a target allocation. I think that’s true of virtually all robo advisors, and many human advisors too.
108.
by cschmidt 4y ago
I find it kind of strange that the products many of us have our money in are so static. I use the Schwab robo-advisor. It re-balances to a target allocation. But moving to high inflation regime with a recession on the horizon, with the f
109.
by cschmidt 4y ago
You can sign up for an openai account here: https://openai.com/api/pricing/ You get some free usage, and then pay per token after that. Once you've done that you can play with chatGPT for free, as of now. I
110.
by cschmidt 4y ago
The continuity of family companies in Japan also comes from the practice of adult adoption, when the family line doesn't have a suitable successor. https://en.wikipedia.org/wiki/Japanese_adult_adoption
111.
by cschmidt 4y ago
This was a very good read: The good delusion: has effective altruism broken bad? https://www.economist.com/1843/2022/11/15/the-good-delusion-... Edit: this article is discussed here https://ne
112.
We are as gods: a film about Stewart Brand(weareasgods.film)
3 points by cschmidt 4y ago|0 comments
113.
by cschmidt 4y ago
That wouldn't really help. Let me explain in a bit more detail. The results depends on the query matrix, which will be different for each set of queries. We have a query matrix Q that of dimension 10000x128. And we have another vector
114.
by cschmidt 4y ago
I hesitate to mention this, because you probably know it and are doing it this way. But some other poster mentioned "and then do cosine similarity". In this case, you're going to want to preprocess and normalize each row of
115.
by cschmidt 4y ago
Huh, at one point you could have multiple queries, but it looks like that is deprecated now. https://www.pinecone.io/docs/api/operation/query/ So maybe it wouldn't work for you use case.
116.
by cschmidt 4y ago
You can fit 1M vectors in the free tier of www.pinecone.io if you want to experiment. I'm not sure how fast having that many query vectors would be. (I'm a happy Pinecone customer, but only use a single query vector.)
117.
Former security chief claims Twitter buried ‘egregious deficiencies’(washingtonpost.com)
8 points by cschmidt 4y ago|3 comments
118.
by cschmidt 4y ago
When I applied for grad school (long ago) they just said it was better to experience another department. The undergrad classes we similar enough to the grad classes that it was good to see a different approach. So if you had a family rea
119.
by cschmidt 4y ago
I imagine they're trying to say you can't have a sub-linear approach, because it takes linear time to specify the graph (going over it once). So the time to "write it down" is the best you can possibly do (for an exact
120.
The Bad Times Are Coming for Startups(wired.com)
2 points by cschmidt 4y ago|0 comments

More ›