yobbo

49 ms·

241.
by yobbo 2y ago
Intel's all time high was in 2000, if I'm reading the charts correctly. Of course the equivalent can happen to Nvidia. Seems almost certain.
242.
by yobbo 2y ago
That sounds a little too strong as a definition. In sports or games, players have no control over their opponents, but they should have full control of their own actions. It is possible to play completely without stress. But being outmatche
243.
by yobbo 2y ago
He invents his own definition of stress, which makes people responsible for their own stress? Seems dismissive of other people's experiences. From what I have seen, stress in people is always caused by consequences from actions that th
244.
by yobbo 2y ago
The difference to RNNs is that the state in SSMs is a linear combination of previous inputs, which makes it possible to parallelise training in various ways. Gates and non-linearities in RNNs allow the state to be "any" function o
245.
by yobbo 2y ago
You could look into the world of Lotus Seven replicas. https://en.wikipedia.org/wiki/Lotus_Seven
246.
by yobbo 2y ago
Looks like "rivets" are moulded into the carbon fibre surface. So it's more like cosplay rather than a replica.
247.
by yobbo 2y ago
Phrasing it as "doing it wrong" misses the mark. They want the name (scrum, agile ...) without paying for it through the reorganization the would be necessary.
248.
by yobbo 2y ago
The manifesto describes an effect or desirable end state . Not a method. Anything you do that fails to bring you this effect is your responsibility. Coaches are only agile insofar as they are helpful. But this also mean you are free to l
249.
by yobbo 2y ago
What about a type of blind interview where neither of the two know the problem or solution in advance. After, their solution is blindly reviewed/graded by a third party. - Input from the interviewer reflects communication ability, cult
250.
by yobbo 2y ago
Theoretically, HMMs are "models of the world" and transformers are approximations of HMMs in the "forward" algorithm. Something seems sus about how the linear projection ended up exactly in the same shape as their pred
251.
by yobbo 2y ago
ozmo.io feels odd. Considering the examples in the screenshots, it seems aimed at some foreign market, but all text is in English and it reads slightly "off". I think most of website design/copy is about creating the appearan
252.
by yobbo 2y ago
> Two of the companies I've worked for have had their CEOs fired/replaced for poor performance. Ah. Like employees.
253.
by yobbo 2y ago
Predictability is defined in terms of probability distributions, and a price X is typically defined to be drawn from something like Xₜ₊₁ ~ Xₜ + N(μ,σ). The purpose might be to quantify some property of μ or σ or something like that. It mean
254.
by yobbo 2y ago
The are various models from textbooks now that seem (or are presented as) too naive to be applied to financial markets, and too slow (eg gradient descent/expectation maximisation) on 1980s computers with "big data". And then,
255.
by yobbo 2y ago
The word in Germanic that is used closest to "engine" would be maschine/maskin/ ... and so on. It is called motor with respect to its effect , and machine with respect to how it works . A machine is called motor to den
256.
by yobbo 2y ago
Not with any known technology.
257.
by yobbo 2y ago
You can choose to have a gap or failure on your résumé. Either makes you unemployable at the types of banks or consultancies considered "safe" in the article, unless you have some sort of remarkable pedigree or network. That type
258.
by yobbo 2y ago
Your future employer, friends and family will know. People in general don't keep track of other random people.
259.
by yobbo 2y ago
Not only this, but failure has consequences socially and professionally. Startup founders are seen a little bit like fortune seekers ("gold diggers") in that they seek fortune without "having to work". Of course it depen
260.
by yobbo 2y ago
If you're learning haskell and want an exercise that motivates monads and more, this is great: https://github.com/lsmor/snake-fury You build a snake game in stages.
261.
by yobbo 2y ago
(x + √(x² + 1)) is a bit like a "soft relu". When x<0 it goes to zero. When x>0 it goes to 2x.
262.
by yobbo 2y ago
Wow. Literally kowtowing ...
263.
by yobbo 2y ago
Knowing syntax is a side-effect of having written a minimum number of LoC, so grading syntax is in effect rewarding experience. Similar to how a large vocabulary is a side-effect of having read a lot. To me, exams are taken in halls, writte
264.
by yobbo 2y ago
In his latest book ("Functional Design ..."), Martin argues that functional programming today is what object oriented programming was in the 80/90s. Hopefully, this is correct. Before everyone learned common design patterns,
265.
by yobbo 2y ago
I don't see the difficulty as finding the problem (schlep blindness) or an idea of the solution. Rather, it's always imagining realistic ways of implementing a solution. In payments, for example, the problem is never "buildin
266.
by yobbo 2y ago
https://kindxiaoming.github.io/pykan/intro.html At the end of this example, they recover the symbolic formula that generated their training set: exp(x₂² + sin(3.14x₁)). It's like a computation graph with a library
267.
by yobbo 2y ago
Sleepsort is like a hash-based sort using sleep(x) instead of a hash function h(x). You can see more here: https://pages.cs.wisc.edu/~siff/CS367/Notes/sorting.html The function h(x) places the item ordered di
268.
by yobbo 2y ago
> That is you can use a rough estimate of 40% No. But it's complicated ... For example, income tax is charged on the employee, after the employer has been charged employer/social fees of 30-40%. And then, sale of most goods are
269.
by yobbo 2y ago
Also, if you know the distribution of the values, you can achieve sorting or pre-sorting by a letting an order-preserving "hash-function" fill a table, so O(N).
270.
by yobbo 2y ago
The goal here seems to be to incentivise bandwidth-sharing, but this is not the main problem. For a "decentralized youtube" to ever make sense, the problems that need solving are how to compensate content-creators/owners, and

More ›