jamez1
jamez1 on Hacker News
That is incorrect, most definitions you'll see of VaR are incorrect. Think about it logically - the purpose is to quantify the tail events you've observed in the past. Practically that would be the lower bound (as the future can surprise you more).
Read the discussion on VaR here if you're interested in the detail:
https://martinkronicle.com/aaron-browns-red-blooded-risk/
It just appears to apply VaR to failure rates, which is hardly a 'wall street secret'.
The authors also completely misunderstand how VaR works - it is the minimum value at risk, not the maximum.
Value at Risk (VaR) [33] captures precisely these bounds.
Given a probability threshold β (say β = 0.99), VaRβ provides a probabilistic upper bound on the loss: the loss is less than VaRβ with probability β
It is actually a probabilistic lower bound on the loss.
So what does the software look like up until those 'annual cycles'?
It's pretty clear from the economic incentives of the companies producing the software, the sheer quantity of exploits found, and the surprisingly low amount of QA that would be needed to catch most bugs that the aforementioned view is correct.
This is a restriction on physical cash, there is no mention of electronic cash let alone cryptocurrency.
Maybe you've got a bad BGP route?
How exactly do we know children with some form of mental deficiency don't spend more time in front of a screen?
Often the operating business is floated and the vendors keep the buildings, this is nothing new and much more endemic than appears.
You're only just seeing an article on this behavior because a high profile tech company is doing it rather than a factory or a shop.
Highly recommend this book for anyone interested in this line of thinking:
How mathematicians think
William P. Byers
It's proof that the process of research in climate science has large flaws, much like the other social sciences. Given they still do things the same way, why would you expect this time it's different?
This is nothing new, most funds treat portfolio construction as an optimization problem, where the objective function is some risk over return metric.
"The Machine" is an optimizer, all they've done is build one that looks at early stage companies.
That's interesting, thanks for the links. I have mucked around with program induction using expression trees but had not come across MIL before.
It looks like MIL would generalize quite well to bayesian networks, a perfect bridge between his past research and what he's depicting as the future of ML research.
I am making an inference here, but I imagine his complaint is with the lack of ability to derive A and B in your "If A, then B" example. His goal is to have a program that would have a model of reality and be able to postulate different models of reality, I don't think it's that unreasonable to claim we don't know how to build that using mathematics yet.
You do make a good point about his agenda though, out of the entire world it's hard to imagine someone more bias to wanting to see causality be the new machine learning topic du jour!
Some stats:
Taxify has had 130k downloads in the last 4 months.
Ola is at 24k for about 1 month.
Far cry from Uber which does that in almost 2 days
It's about information entropy, you want to maximize the information each character has. using 1000 digits to show what you could in 500 and a header saying "(in thousands)" maximizes the entropy
They pay them more for waiting, effectively moving the working capital up the supply chain where it accumulates to where it can be used.
What framework handles the primary key data type for you? And what frameworks in general do you think were any good in the year 2000
Having a distributed money supply makes it unstable, since you've tied an incentive to creating bitcoin.
Money supply would then be equal to how many miners there are without adding a difficulty, and the amount of bitcoins would go through the roof. Difficulty was clearly added as a solution after this to keep the money supply stable.
Note I didn't say it's necessary to waste electricity to have a stable money supply, only that the reason Bitcoin does is because it keeps it's money supply stable.
High electricity consumption is the trade bitcoin made in its design for a stable money supply, it's not an inherent problem with cryptocurrency.
Make sure you load up on mathematics, in both pure and applied. Try to find work where you'll either be immersed in statistical models or have the freedom to experiment with them.
Capital markets such as equities trading will probably be the best place to look for work experience.
What does the number of publications have to do with anything? One paper can change an entire field, while another hundred could be just filler non-sense to get a PhD with.
Congratulations on your project Oliver.
How far off do you think it is from a regulatory perspective before you can have the cars drive without anybody sitting in them?
That's 1000 cases when we restrict children from taking aspirin, it could be a lot more once we let them
This isn't about eigenvectors in general, it's about a single eigenvector that Google built. The title is accurate, their search product is giving you access to query the eigenvector
> Worse, NHST is often taken to mean that any data can be used to decide between two inverse claims: either ‘an effect’ that posits a relationship between, say, a treatment and an outcome (typically the favoured hypothesis) or ‘no effect’ (defined as the null hypothesis).
This is the whole problem, the rejection of the null hypothesis is not just a ritual to follow before you then accept the alternate. Fisher's core idea is that it only takes one piece of evidence to reject something, while it should take many to confirm something.
So the only conclusion you should draw is about what you've rejected. You can't then take that and infer the acceptance of something else. It's a subtle but important distinction that is the whole reason for this process.
It's a great theory but how do you know that education causes lower fertility, and isn't just correlated to lower fertility?
I'd wager getting out poverty causes both education and fertility to improve and that is the real driver in this study.
It states that they create a hash server side, why couldn't that happen client-side?
> it's not implemented in Javascript.
They can implement it. Hell half the readers on this blog could implement it..
Imagine you are a climate scientist, and your mission is to save the world by working out how we are hurting it.
You constantly run and re-try computer models until you find one that you can publish.
After 1000 different models you build one that seems to say there is some catastrophic effect caused by by humans with 99% confidence.
The model then gets published, adds to the growing body of climate models that were made the exact same way your one was.
Tell me - do you not see what is utterly incorrect with the above process?
It depends what your null hypothesis is. Your contrarian friends only need one piece of evidence to disprove the climate catastrophe narrative.
Whilst you've taken the worst case as your default assumption, which is why their train of thought seems off. You sound like you need a large body of evidence to disprove climate catastrophe.
When you start to enter the realm of modelling, especially of complex systems with feedback loops, you aren't really conducting science any more.
Unfortunately the brand of science is used to promote these models, which is disingenuous. These models aren't like physics models, they are like economic models.