cosmic_ape

How did it happen that default funds of the clearing houses had to be accessed? Did the spread change in one day from normal spread to 17 times normal? Otherwise the margin should have depleted only his own money.

the page doesn't load

I certainly agree re:high dimensions. This is because to integrate, you don't need to approximate the function everywhere.

Just to make sure, the uniform grid on n points will also lead to O(1/n) convergence, right?

I get that the uniform grid is not an open sequence, just want to understand which kind of convergence do you have in mind. For which class of functions?

Another happy DDG user here. But I always wonder - DDG is basically an aggregator, it doesn't have its own general index. Isn't there a danger that Google and Bing will just close the faucet to it one day?

Just to remove any doubt, I have a lot of respect for nurses.

But the question was more about, say, should we close now all the MD schools and replace them by the alternative path you describe?

Imagine a hospital where a new initiative is taken. Nurses follow doctors during their daily patient rounds for 6 months. Then the nurses are expected to do the rounds instead of the doctors. No doctors are left on the premises.

Its not very clear cut, there are some blurry boundaries, but still, would you want people to use that hospital?

I hoped to see there an elaboration of what a "Japanese Dream" is, but there isn't much about that in the article.

The particular behavior you mention can be modeled by a Semi-Markov Model. In these, the time you stay at a given state can be an arbitrary fixed distribution. In a regular chain this time is geometric, so memoryless. But other distributions, with Gaussian like tails or power law tails would produce the effect you want.

I see, thanks. I think they should have just said distributions are skewed, and skewed differently, if thats what they mean.

Can anyone give an example of how the "common view" could be true at culture-level but not at individual-level?

This editorial spends a page and a half talking about it without a single example.

There are other methods, no doubt. But last time I was involved with the field, software verification was still an experimental thing, that kind of worked in some niche cases, while SAT based verification was applied to every new arithmetic hardware design. Not to disregard other methods, but I'd say SAT still qualifies as being _the_ component.

Its not perfect, but optimization problems are usually solved with SAT as follows: Suppose you have some X that you want to maximize. You encode a constraint X>=5, and see if there is a solution to that. If there is, you try with X>=6. You continue until you can not find a solution or get unsat.

In your case, X would be the number of "nice to have constraints" that are satisfied. It would be simpler to formulate the these conditions in some SMT language, rather than in SAT directly.

For those hearing about SAT solvers for the first time, its worth mentioning that they are the basic component of hardware (and software) formal verification systems.

This is probably the largest practical application domain, which also drives most of the academic development.

But keep in mind this, as another comment here mentions:

> Accustomed to meeting the highest of standards, he saw his dissertation as mediocre. Quietly, Venkatesh started eyeing the exit ramps, even taking a job at his uncle’s machine learning startup one summer to make sure he had a fallback option.

yep. That was my point, $59,000 seems too high.

According to the source given in the article [1], median worker wage is 44,564 (for a 40-hour workweek).

edit: It could get to 59,000 if these people actually worked 53 hours a week.

>>>According to the Bureau of Labor Statistics (BLS), the median wage for workers in the United States in the fourth quarter of 2017 was $857 per week or $44,564 per year for a 40-hour workweek.

[1] https://www.thebalancecareers.com/average-salary-information...

The description says:

>>>This list focuses on notation as "a series or system of written symbols used to represent numbers, amounts, or elements in something such as music or mathematics." This is distinct from a language (computer or natural), interface, diagram, visualization, or tool, but may overlap with them.

Not sure how this is distinct from a language. She has an example of a notation for knots. This is a way to describe knots. Now, consider for example a notation for describing binary valued functions of a binary arguments. We have tables, the boolean algebra, with DNF and CNF, decision tree representation, and for larger functions - VHDL, for example. All are languages.

In fact, distinguishing between "language" and "notation" looks like a good example of "bad notation".

I don't think there's a contradiction. My comment was about what it means to understand the model, or perhaps about what it means to do science, but not about what is necessary or sufficient for applied work.

But if we do talk about that though. Yes, science proceeds by modifying or sometimes totally abandoning the assumptions. No silver bullet, and the point is to assume as little as possible.

In general, if things "just work", its not a reason to abandon attempts to understand them. Things "just work" until they don't, especially in finance.

Consider two guys. Guy A, he predicts that the sun rises every day and is there until dusk. It "works", and so, as the joke goes, he does not touch anything. And guy B, he knows physics and has telescopes. And he can predict the solar eclipses. He knows the failure modes of the model of guy A. Consider the difference between the two.

Its funny, and characteristic of that part of the field, that mathematical provability in not even mentioned explicitly.

A model (and a learning algorithm) should be interpretable if we can clearly state the assumptions, and prove that under these assumptions, we get what we state we get.

We can predict the movement of the planets (short term). This is interpretable, because we only assume a model of 3d space, and Newton's laws(1). The rest is mathematics, which gives us elliptic trajectories, etc. The resulting predictions may be wrong, due to wrong assumptions, but the model is still interpretable.

Linear regressions, histograms, decision trees(2) are all interpretable. We know exactly what they do, under proper assumptions. Yes, these models can be sometimes manipulated. But we know this precisely because we know exactly what they do. Whether we should use these models, or statistics in general, in courts etc, is an ethical decision, and perhaps practical decision, similar to whether we should use the death penalty. Has nothing to do with interpretability.

With CNNs, at the moment we have very vague understanding of both the assumptions and the models. When we do have this understanding, we will likely also have better and simpler models.

Mathematics was always a tool for understanding reality. Proving is understanding. And it was pretty successful so far.

(1) and some other things -- the rest of the planets are far enough, etc. (2) given enough data wrt the size of the tree, under proper distributional assumptions.

Its an issue. But if all the code is open and audited, there should be a way to ensure only aggregates are used, and the reports are anonymous. There would also be a related issue of bot accounts who would try to sabotage the statistics for fun, but should also be reasonably solvable.

For offline navigation there is data, the OpenStreetMap, for most places on the planet. There is no very good app though, which would have seemed to be the smaller problem.

For real time navigation, Waze type solution could work -- aggregation of real time data from users. Also, smart city projects have open traffic data APIs, although not sure wether there is something properly real time, or it something like hourly aggregates.

So that would be still a giant project, but data does not seem to be the bottleneck.

>>Employers will also have to be more open to attracting international talent to fill the jobs created outside its oil and gas sector, she suggests. On the other hand, young Norwegians “used to being able to work wherever they like” may need to become increasingly “focused on where the skills are needed”...

So she means that while the population is well educated, many do not choose the "right" education?

Adam may be not the clearest example. I'd say the clearest example is the original GAN paper, where "theorems" where completely unrelated to what they were doing and trivial at the same time.

>>and even if it risks publishing an accidentally incorrect proof.

No. If a researcher has even a slightest doubt about the correctness of their proof, they should either spend the time to verify it or talk to an expert. There is no lack of experts in convex optimization in particular. Or, instead of publishing the proof, one could just say "we believe this is provable by these methods". But one can not try and shift the burden of verification to the readers, while claiming the results. This is not how trust is built.

I should say its very easy to construct synthetic datasets on which it is very clear that the only role of large layers is to supply more opportunities for the initialization to get it right.

So, not much surprise there. But they claim they can extract the smaller subnet, which could be useful. Except they only provide experiments with very small nets so far, as the comments on reddit point out.

There are books that are full of bells and whistles. And there are books where the writing is not necessarily most polished, and does not follow the most recent hypey style, and they still easily thrive for centuries on the shear power of the ideas inside. Dostoevsky and Phillip K. Dick are probably the standard examples. They give the reader something of value, something that allows the reader to process the world in a new, or additional light.

I do not know much about the databases, but it I'd imagine its one of the backbones of the information age. In particular probably the efficiency aspects. It is a powerful material. And so as long as you come into the classroom and make it clear that today you are going to deal with something that makes the student more powerful, the students are going to listen, no matter which manner of delivery you choose. It may not always seem like this, but they are young adults, not little kids whom you need to trick into eating by stories or something.

Being an industry veteran, you likely have a great perspective on what works and why. The usual things, examples, critical distinctions of what could be a good idea and what could be bad, are useful. But whether its eventually provided as HTML or not, I don't think it matters.

A bit different, but something similar does happen already, when people rate taxi drivers, cashiers in a store, or people commenting on airbnb hosts. These reactions often get personal rather than being based on an objective evaluation of a transaction.

Point is, the "flight crew"s and even the "pilots" are not the corporation. The corporation may easily sacrifice their privacy if its believed to contribute the bottom line. Even the CEO might be sacrificed. The really private things about a corporation are its finances, its tradeoffs of customer benefit vs "efficiency", its market segmentation policies, etc.

and not enough coupling arguments.

I agree with what you mean.

But do you read all the medical literature before going to a doctor when something just feels weird?

Or, somewhat differently, do you read the source code of all packages on your computer to ensure there is no malware?

I'm a universalist as much as I can, but there's a reason specialization, division of labor, results in productivity increase. Somewhat funny that Kant himself also argued for division of labour, just a year after publishing the Enlightenment article, [0].

[0] https://en.wikipedia.org/wiki/Division_of_labour#Immanuel_Ka...

That's useful, thanks. But to be a bit pedantic, this can only show that at a particular place where the measurement is taken, the earth has some curvature, that its a bit like a hill there. This does not show that the earth is globally a sphere, which would mean for example that if you go in the same "direction" long enough, you arrive at the same point.