Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
yobbo
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
yobbo
8d ago
> Has what students are expected to know changed since the 60s? No, basic algebra and calculus has not changed.
2.
▲
by
yobbo
21d ago
Yann Lecun is "famous" for (besides being one of the fathers of "modern" AI) his opinion that transformers are inadequate and other directions should be explored.
3.
▲
by
yobbo
23d ago
As always and everywhere, it's who you know (and who knows you) that matters.
4.
▲
by
yobbo
24d ago
Yes. It's like looking at the (Apollo) moon rocket launch and then suggesting teenagers should learn to build rockets in their garages for the coming space age. It is viable as a toy project, but there are vanishingly few career opport
5.
▲
by
yobbo
1mo ago
For games in general, one measure of complexity is branching factor. It means average number of possible actions or states at each turn. It is knowable. "Solvability" would mean number of turns to solve the game. It is known for s
6.
▲
by
yobbo
2mo ago
Stuff like this exists for bicycles. For mopeds and motorcycles there are electrification kits. It should be possible to just replace the flywheel altogether with a generator/motor that both brakes and powers during each cycle, cancell
7.
▲
by
yobbo
3mo ago
This (classic) argument is symmetric with respect to the value of money and quantity of goods. As in "if you know money will buy more in the future, it increases your incentive to sell now rather than wait for higher prices. And if you
8.
▲
by
yobbo
3mo ago
4) add taxes to effectively double price of gasoline
9.
▲
by
yobbo
3mo ago
Normal use means loading data into the GPU for each batch. The load is not even, though training might be worse than "production".
10.
▲
by
yobbo
3mo ago
Looks very nice, but I can't find numerical gradient checks, which is helpful when verifying that backward pass is correct: https://github.com/markusheimerl/gpt/blob/main/transformer/a...
11.
▲
by
yobbo
4mo ago
Testing was the annoying flood barrier. AI is the rainstorm that shows why it was necessary.
12.
▲
by
yobbo
4mo ago
The point of automation is that it reduces the bargaining power of human workers, but they still have to trade their time and effort for wages. When automation improves, the wage proportion of the created value shrinks.
13.
▲
by
yobbo
4mo ago
Housebuilding is motivated as much by need for housing as the prospect of increasing market value, "building equity" and so on. This prospect depends most strongly on location, which means the land price dominates the cost of cons
14.
▲
by
yobbo
4mo ago
> Sweden had nothing like this Not entirely true. AstraZeneca and ABB are examples that remain partly Swedish but many companies were merged into big multinationals and eventually marginalised.
15.
▲
by
yobbo
4mo ago
My understanding after scanning the code examples is the technique expands the dimensionality of each data point with a set consisting of the quadratic coefficients of its existing dimensions. I thought it sounded like kernel PCA.
16.
▲
by
yobbo
5mo ago
If your goal is to translate Haskell (or other garbage collected code) pattern-for-pattern into Rust, you will almost certainly burn out.
17.
▲
by
yobbo
5mo ago
> difficult or impossible in Rust were to me pretty basic patterns for modularity Many things are plainly not permitted, either because the borrow-checker isn't clever enough, or the pattern is unsafe (without garbage collection and
18.
▲
by
yobbo
5mo ago
The reason for exp(x) is that its derivative is exp(x), which makes it possible to express the gradient of s(x) in terms of s(x), or both in terms of exp(x). This simplifies the computation of backward pass.
19.
▲
by
yobbo
5mo ago
In ancient times, slaves didn't revolt because they were oppressed by weapons and violence. Instead of paying workers, resources were used to pay and feed the military. Slaves were considered "war bounty" and tradeable goods.
20.
▲
by
yobbo
5mo ago
The point is rather that in a pure language, each io operation needs to be dependent on a sort of "world state" which is updated for each operation. They chose to implement this state as the io monad but there could have been othe
21.
▲
by
yobbo
5mo ago
> I don't know if I agree with either assertion… Yes, I don't think this matters. Much of "knowledge work" was always a proxy for something else. High quality in terms of typos and errors is mainly a signal of respect
22.
▲
by
yobbo
5mo ago
> this law will make phones _worse_ for most people Not really. The battery just needs to have a connector rather than soldered, and no other things blocking the battery once the back-case is opened. Realistically, a service shop will do
23.
▲
by
yobbo
5mo ago
They are conditions to be met. It's not enough to proclaim them as "your process" and expect results. When playing piano, the condition you are measured by is acoustic harmonies in the air, not finger movements. The only reas
24.
▲
by
yobbo
5mo ago
Deepmind was not a viable business at the time Google acquired it. Today, it is probably even less viable. It functions as R&D-lab for Google which has its own products and datacenters.
25.
▲
by
yobbo
5mo ago
Start by shifting taxation from worker incomes to corporate incomes?
26.
▲
by
yobbo
5mo ago
"What you do" seems more sympathetic than "who you are" and "who you know". American culture might be more meritocratic at a basic level.
27.
▲
by
yobbo
6mo ago
"The only reason people disagree with me is because they are emotionally deficient."
28.
▲
by
yobbo
6mo ago
> No one has ever made a purchasing decision based on how good your code is. I routinely close tabs when I sense that low-quality code is wasting time and resources, including e-commerce sites. Amazon randomly cancelled my account so I w
29.
▲
by
yobbo
6mo ago
> We can assume they are doing so at a profit This is false. We may assume it's the most efficient way of generating revenue given their GPUs, but their overall profitability will just be a guess. They would still have incentives to
30.
▲
by
yobbo
6mo ago
> Isn't that how LLM models are trained right now It's neither how computer chess works or how LLMs are trained. Computer chess uses various tricks to prune the search space of board states, where the search is guided by the &q
More ›