ford

6 ms·

31.
by ford 1y ago
This sounds like specific instance of what I think most people believe about AI: It's good for tedious or well-scoped tasks, but shouldn't handle things that are core to your job. I think students see this (use AI as a friend&#x2F
32.
The Last Question (1956)(users.ece.cmu.edu)
1 points by ford 2y ago|0 comments
33.
by ford 2y ago
"Law" might not be the right word - but there's no denying it's scaling with compute/data/model size. I suppose law happens after continued evidence over years.
34.
by ford 2y ago
Seems like both: - AI Labs will eat some of the wrappers on top of their APIs - even complex ones like this. There are whole startups that are trying to build computer use. - AI is fitting _some_ scaling law - the best models are getting be
35.
by ford 2y ago
Given the pace of AI lately - might be worth mentioning this is from March of this year
36.
by ford 2y ago
The other problem is IDEs are incentivized to build their own AI coding tools instead of letting other people plug in
37.
by ford 2y ago
How bad of a sign is it that so many people have left over the last 12 months? Can anyone speak to how different things are?
38.
by ford 2y ago
It's a negative signal not to be able to critically assess how you're doing - both for an individual and a company. The best teams know that being embarrassed about problems, or, at minimum, acknowledging some of the downsides of
39.
by ford 3y ago
Jay Alammar has my favorite sequence of tutorials from basic neural network math to GPT2. Particularly [0], [1], and [2] [0] http://jalammar.github.io/illustrated-transformer/ [1] http://jalammar.github.io&#
40.
by ford 3y ago
My pet peeve in college was people who "studied" for "12 hours yesterday" but never had more than 12 minutes of continuous focus. Personally I have the most success with a couple of hours of focus followed by 30-120 minu
41.
by ford 3y ago
This made me wonder why it's called programming (since clearly it's not the sense of the word programming most HN'ers are used to). https://en.wikipedia.org/wiki/Mathematical_optimization#Hist...
42.
by ford 3y ago
Software engineers interested in ML/algorithms should learn about linear programming. It's surprising how many problems can be formulated as linear optimization. For example, in college I was talking to my Industrial Engineer frie
43.
Ask HN: What are your favorite end of year reviews?
1 points by ford 3y ago|0 comments
44.
by ford 3y ago
What's confusing about how having more money might let you hire more people to build more things or build things faster? If something is an obvious winner - why handicap it by limiting investment to what it can make on its own? What if
45.
by ford 3y ago
Do you have more data about why they are useless? Safes are still useful even if they have never tried to be breached, and armed forces are useful even if a country has never been attacked. Perhaps the fact that the TSA exists means fewer c
46.
by ford 3y ago
The advice mostly still applies if you replace "Stanford" with "Higher Education Program" For most 18 year olds, doing well in school has been _the_ thing to focus on, for their entire life. Using school as a means to ac
47.
by ford 3y ago
There's very little incentive to write headlines that aren't clickbait
48.
by ford 3y ago
So far it's been ~8 months since ChatGPT started the (popular) LLM craze. I've found raw GPT to be useful for a lot of things, but have yet to see my most frequently used apps integrate it in a useful way. Maybe I'm using the
49.
IBM Suggests their Quantum Computer will have an application within 2 years(nature.com)
1 points by ford 3y ago|0 comments
50.
by ford 4y ago
There's an interesting short story "Story of Your Life" by Ted Chiang that is related to this concept. It was the motivation for the movie arrival (though IMO is much better - the movie adds some drama that does not exist in
51.
by ford 4y ago
Perhaps this is another of the early (?) nails in the coffin for traditional higher education. If it becomes harder to assess if someone learned something (with a grade), the results of that assessment (GPA) become less valuable. Software h
52.
by ford 4y ago
I see where you're coming from, but just because mice don't live as long as humans doesn't mean all biological processes happen 16x as fast. They don't defecate 16x as often as humans or sleep 16x per day. Do you think t
53.
by ford 4y ago
+1 boredom isn't quite right. I'll check out that book.
54.
by ford 4y ago
Is there more research on how boredom relates to creativity in general? I've been wondering recently about how my default in most cases is to look at my phone (reddit, social media, texting). That must detract from my net new ideas. I
55.
What in the Ethereum application ecosystem excites me [Vitalik Buterin](vitalik.eth.limo)
1 points by ford 4y ago|0 comments
56.
by ford 4y ago
My hypothesis is that few companies are big enough to be able to dedicate many people to _exclusively_ modeling. If a modeler has 4 weeks to spend on learning new things, odds are most companies would benefit more from the modeler learning
57.
by ford 4y ago
I long for the day I can carry my phone or watch without needing keys or wallet. The fact that most adults carry at least 3 things (keys/phone/wallet) at all times in 2022 is crazy.
58.
by ford 4y ago
I've done some work in this area and have indeed found that simpler statistical models often out perform ML/DL. This is true for single time series, where we are predicting P(x_t+1 | x_0..t) DL has advantages when you a) have addi
59.
by ford 4y ago
A friend of mine has had a similar experience. They started the process in July, did onsites in August, were told they weren't filling roles for a couple of months, and recently started team matching; but it is slow going.
60.
by ford 4y ago
Generally problems that require more bits are "harder"

More ›