nicolapede

What's your workflow/tool set?

> But allowing calculators WILL make those who like math reach much deeper into the field than without.

Have you ever run into any mathematician that praised the calculator for his/her career? I’d be really curious to read about that.

No. Just look at equations 6 and 7 in the link. The expected value of the move can be either positive or negative depending on the model parameters.

If you look at the Baltic Sea before and after the invasion of Ukraine, that was largely a neutral space (before) that has since turned into a NATO one -- excluding only the Russian exclave of Kaliningrad.

I would argue this makes any further move by Russia in that space quite problematic.

Thanks! Those look exactly what I need! I struggle to keep my hands warm below -3C.

The curious thing about this, though, is that of course you can start by replacing software engineers with it, but how far away are you going to be from being able to replace a CEO with it? I would say not that far away.

It does feel magic indeed.

Nice write up, thanks for sharing it. Would you know of any introduction to back-propagation written in a similar fashion?

The day AI can be used in war, through robots for example, loosing a war might mean total annihilation. That is beyond nuclear weapons risk.

If anyone likes to check the clarity of his view, I suggest "The abolition of man" -- chapter 3 in particular. I find it so prescient of modern times in so many ways. Astonishing.

Side note, if you want to explore your inner experience in a more active way, try finding a local meet up that does shamanic drumming with a guide. I know this sounds weird, but our subconscious is all about symbolism, and if you can play along it's basically a short and gentle drug-free psychedelic trip in a good setting that can be very rewarding.

I am really intrigued by this (and I don't find it excessively weird). Would you have any material to recommend if one wants to understand how it works and why before actually trying it?

Matplotlib has been replaced by tikz [0] for most of the figures I need to generate (they would feed LaTeX docs anyway).

[0] https://tikz.dev/

I agree -- there is a lot of variability depending on boroughs. Hope they will extend the good practices more and more.

As a regular commuter who cycles through Surrey and south London, I must say that most cycle lanes in London are a joke. They are either shared bus lanes (the good case, albeit you might get in trouble when multiple buses queue one after another) or half-a-meter wide ones, where cars are allowed to cross your lane from both sides. Sometimes using the car lane just seems safer than that.

Is there any analysis explaining how the new interest rates regime is affecting the number of employees in tech?

I think the impact on VC is probably more direct, but, given what is happening all over FAANGs, I would imagine there could be some causal relation there, too?

It was a great final, exciting and fun to watch, at least towards the end. As a World Cup edition, though, I found it a disappointing one. It might be that the unusual season when it was played made physical skills more important than usual, not sure. It was difficult to find good, modern football though (beside maybe with Brazil, but they self sabotaged them in a few minutes folly). In general, it seemed that the whole world wanted Messi to win and we eventually had that.

The fact that the two (possibly three) top players of the tournament played for the same club is something to ponder about I feel.

What are the new language features that let you avoid using smart pointers?

That is a good point, thanks. It is not that I want to change device only because of lack of support for open ebooks. In general I’d like to start being less reliant on Amazon and I just thought of asking for advice starting from this discussion.

I must say 8 inches does sound interesting. Thanks!

Thanks I’ll check it out!

Thanks - that is amazing! As a Kindle Oasis owner, which I love in terms of screen size (7 inches) and auto adjusting light, is there any non-Amazon e-reader device alternative that anyone would care recommending?

Actually, you could say we're in the middle of a transition.

I recently noted the same while watching Severance. Almost all of the episodes, barring one or two, were directed by Ben Stiller. I found it one of the best (first) series ever.

IF someone can pull that of while protecting the individuals I believe it's alphabet. Together with deep mind and all the other research they do.

What makes you think they will protect the "individuals"? From a privacy persepctive, I would say they have no clear incentive in doing that, given individuals are the product they sell to companies that need to push ads into society.

But many of the problems that remain untouched by DL, like human cognition, are NOT driven by gradients.

This seems to me quite a deep insight. But how would you formally define a process that is gradient-based?

Alpha Centauri is one of my all time favourite game. Out of curiosity, does anyone know if it is possible to replicate the Human Hive strategy from Alpha Centauri in other Civ games?

Maybe someone already said it, but I just don't see Twitter as a public space -- it is not Hyde Park Corner, it is a large courtyard, very large indeed, but privately owned. From this perspective, I am really not bothered of who is on it and who is not (I am not on it).

The topic discussed in the Twitter thread, on the other hand, is much more interesting. I am a layman in ML (although I have implemented a NN myself, but I am not a data scientist) so I offer my view hoping to get some feedback and be corrected. My view is that the original post was correct and LeCun was not. I just can't see what sort of objective function you can use to solve every potential problem arising in the future about similar issues. You want a "totally unbiased" model? Then, you use total diversity in your training set? Can't it happen that the model will then enforce diversity in a non-diverse picture? Is that your ideal result?

That might be a silly example and maybe not 100% correct for the example discussed, but, in general, my layman feeling is just that ML cannot help in cases where moral judgement is required. It can provide a good tool in any other case where, for example, a quadratic error is a reliable error measure, but that seems a hard limitation.

Given that it can take N weeks to die, wouldn't it make sense to calculate the mortality rate at a given time t as the number of deaths at time t divided by the number of reported infected people at time t-Nweeks ? (I am using N =3 for my back-of-the-envelope calculations).

It depends what you mean by "getting a better grip". There are books on scientific topics that do not rely on technical details. When they are great, they are so exactly because, even with this constraint, they manage to clearly convey the elemental notions to a layman ([0] is a great example). It is debatable whether the grip you get in this way is better or not. Certainly it can get deeper, when complemented with the right analytical tools.

[0] https://www.amazon.co.uk/Relativity-Routledge-Classics-Bertr...

Curiously Recurring Template Pattern (CRTP) [0] in C++.

  class Derived : public Base<Derived>

We talked a lot about how literally no one on earth is capable of seeing an understanding your flaws as well as you, but also most humans on earth are infinitely less interested. The lens you view yourself with is totally unrelated to the lens with which the rest of the world sees you; you'll drive yourself insane if you think you and the world interpret you the same.

What was the lesson you learned specifically from this?

May I ask you where you ordered them?