salty_biscuits

But they have

https://en.m.wikipedia.org/wiki/Jaguar_C-X75

Turbines get more efficient as they get bigger.

Depends on if the result is due to skill or luck?

You can get 80ish percent with a linear model (if you one hot the labels).

I remember a fairly sketchy bit of gear at my old university that an amateur could have definitely made for themselves (if I recall correctly the dish was made with chicken wire). They used it for observing the Vela pulsar for a very long time (like years and years) and looked at how the pulses changed with time.

Yes.

I might not be able to give a super coherent picture but I will have a go. The force that hold nuclei together is based on the residual strong force, it is a complex system so the models used for it tend to be semi-empirical (yukawa or reid potential). They go from being repulsive at short distances to being attractive with something like an exponential decay as you go further away. The force is very strong at 1 femtometre and pretty much negligible 3 femtometres away. This is only the residual of the strong force though, which is what you can think of as holding quarks together. This force is very strange in that it doesn't get weaker the further the quarks get apart, but if it gets far enough then the energy will create new particles. That's just the strong force, I will leave it to someone more knowledgeable than me to try to explain the weak force.

It's just a preprint. I appreciate the fast update of progress on a new development. Just don't give it the same weight as a peer reviewed piece.

If you are really close to one pole of a magnet and far away from the other then the force scales like 1/r^2, it is in the far field under the influence of both poles of the magnet that it scales like 1/r^3. Electrostatics scales like 1/r^2 (because you fet isolated charges). The other two are a complicated story.

Algebraic versus differential if that wasn't clear

You can write out the equations of motion for a bicycle that will very accurately predict the dynamics. You can put these equations into a numerical simulation and predict motion very accurately. You can change the parameters of the model and do simulations with high confidence. Just because there isn't some neat little equation that says exactly what each parameter change is going to do (without doing the simulation) doesn't mean that we don't understand bicycle physics. It's a silly line of reasoning. Those articles are hyperbolic.

The mineral containing Zinc is called Sphalerite, which is based on a greek etymology for "deciever" because it commonly occurs with Galena (which has Lead and Silver) and looks similar.

I had it in my head that the lyapunov exponent for the solar system is approx 15 million years.

I was grinding through Finnegans wake. Took ages! Dropped my book on the floor of the train one day and the bookmark fell out. Worst day of my life, I couldn't find out where I was up to and just gave up and stopped reading it.

"Vanilla deep learning models are statistical models (a la linear regression) and not probabilistic models (a la Gaussian mixture). It is important to maintain the distinction."

Isn't this just a matter of interpretation of the models? You can interpret linear regression in a Bayesian way and say that the prediction of the linear model is the MAP of the mean, you can also calculate the variance, the l2 norm objective is saying the distribution of errors is normally distributed, l2 regularisation is a normal prior on the coefficients, etc, etc? All the same stuff can be applied to deep learning models.

Maybe I don't understand your distinction between statistical and probabilistic though?

To be fair though, there is a wide spectrum of "usefulness" for pure mathematics concepts in applied mathematics. For less contemporary examples you don't need to know about Lebesgue measure to do practical numerical quadrature for applications. Or you don't need to know that rotation matrices are part of the SO3 group to do computer graphics. Sometimes the abstractions are powerful and sometimes they are kind of a distraction (or even make communication a problem). There typically needs to be a super compelling application to motivate a shift in baseline knowledge.

See a lot of pessimism here, and heard a lot of the same stuff by grumpy engineers over my career "nobody knows how to do statistics properly", "scientists are bad at coding" and "managers don't care about quality and rigour". I'm old enough to say it all pre-dates data science as a term. It all comes across as a conspiracy like everyone is bad on purpose.

There is a null hypothesis here, that the average person in role x is just average at that role. It is an extraordinary person who has high level skills across multiple domains like maths/science and coding, maybe so extraordinary that they wouldn't be working with you...

I'll admit that the article rings true, but I think there is an implied intentionality that I don't agree with. We are all just plodding along, doing our best with limited information and skills.

I have high hopes for this project

https://github.com/JuliaLang/JuliaSyntax.jl

Isn't the historical progression cybernetics turns into control theory (via Norbert Weiner) and control theory turns into reinforcement learning. So just had a few rebrandings and never went away.

Yes, you can think of so many methods of signal analysis as searching of the right set of overcomplete basis functions that give you a nice sparse representation for your problem (machine learning included).

Partially true. You can get away with a more efficient thermodynamic cycle on the ICE due to the more uniform load on the engine so they can be more efficient.

Autocorrelation is a delta function. Basically a random known signal, so you can disambiguate simply from background noise (look into matched filters).

Plenty of counterexamples to the claim

https://mathoverflow.net/questions/25630/major-mathematical-...

I don't know, plenty of alienation in 19th century literature where everyone was living in everyone else's pocket. A lot has changed in that time, including our openness to expression of our internal suffering. I grew up in a similar way, my overwhelming memory was that it felt utterly stifling! I couldn't wait to move away and be anonymous.

Social/cultural thing. Can't leave before the boss leaves. Have to arrive before the boss arrives.

Time delay embedding (i.e. the observed value at different time lags as a feature) is the usual trick to turn time series data into a tabular form for this sort of regression.

My experience with "flat" structures in the past has been the lack of an explicit hierarchy created an implicit hierarchy, which caused compensation to not match actual role and some diabolical politics. Never again. I've kind of moved on to wanting to work at places that run on information and not confidence instead. You can work out which type of place you are at pretty quickly by talking to upper management. Unfortunately most people want confidence and not information.

The subways in Japan are fantastic. Interesting that they make a loss on the trains themselves and more than compensate in profit from retail and real estate.

Agree, I just think the modern version of money makes it easier to hold the two ideas in your head that "I'm a good humanitarian" and "they can't afford a house because they are bad with money". Back in the old days they just plain thought poor people were inferior so they deserved poor treatment because of that.

It's really old and has a lot of history post ancient Egypt a fair bit of it abusive, e.g. it looks like the sphinx's nose was chiselled off (not shot off with a cannon by Napoleon in the myth). People used to climb all over the great pyramid, etc. It's cool to go to the more recently discovered places like the valley of the queens to see the paint and hieroglyphics. All the stuff near Cairo has had more human interaction.

I think the nuance is that money is made up, not that it is imaginary. It is an abstraction that lets you live in a society that will let people be cold and hungry without feeling too bad about it.