Not zero but I would expect someone to just buy the name from the liquidators instead. Say if Home Depot decided they wanted to expand the furnishings/decor side.
Hanlon's razor never applies when politics or money is involved. Best to just suspend your judgement instead of giving politicians the benefit of the doubt.
This is explained in the talk he gave. One of the benefits is that with new syntax the semantics can change without breaking backward compatibility, e.g. [[nodiscard]] can become default for all functions written in the new style.
My impression from reading about chess engines online is that drawing is far simpler if that is your explicit goal. Several comments online on different websites claims that they are able to end games in draws against stockfish.
There's a parameter called "contempt" that you can increase to make the engine avoid draws.
The idea is that if the engine has "contempt" for the opponent, it assigns a negative value to a draw, instead of assigning it a value of zero. The engine will then only attempt to draw if it thinks that it is actually in a worse position, but it will not go for a draw in an objectively drawn game. The engine will even play suboptimal moves in order to avoid drawing.
When playing against evenly matched opponents, contempt makes the engine perform worse. But against weaker opponents (such as human super-grandmasters), contempt can make the engine score significantly better.
There are various videos you can see of strong players forcing draws against stockfish (or engine cheaters online etc.). The general way to do it is to trade everything but rooks and then force a very closed position. Then you shuffle for 50 moves or until contempt makes the computer sacrifice material.
I mean I can force a draw against stockfish by just using stockfish myself to beat stockfish.
Look to put this into perspective back in 2008 Hikaru Nakamura a world top 10 player won a game once against the then top engine and it was a significant story in the chess world that day. It's been 14 years since then. If people can easily draw stockfish I'd like to see them do it live giving stockfish the same clock.
One problem with trying to draw stockfish by trading off its pieces is that stockfish is so dominant it tends to win more games the more complicated positions get, so it's actually hardcoded to be bias towards making positions as complicated as possible, so it will resist you closing the position and trading off pieces to an extent. It's also non-deterministic. You might be able to force it to a draw in some games but it's hard to do this reliably.
I'm also by the way not saying Magnus cannot draw stockfish 1/100 games, I'm saying it wouldn't be easy.
"Magnus Carlsen would have a real hard time drawing stockfish on a phone 1/100 games." is what you initially said.
If you're running a fully decked out stockfish with classical time controls, yeah the human will lose every game, but if you aren't, then humans can win in 15 second time controls [0], where computers have a distinct advantage of not needing dexterity to move the pieces, and against online cheaters[1], and with some prep (which like, you get against humans too!) they can win with weird prepped lines (Jonathan Schrantz does this fairly regularly, and he's not even titled).
And the systems those players beat are, uhhh, stronger than a phone. Edit: to substantiate this I just checked, stockfish in the TCEC, which is the big beefy 3700 elo one has hardware to evaluate 200 million positions per second on average. my phone sustained 350k/second, although that was in browser and droidfish is probably 10x faster. But you're still looking at 100x slower than the baseline. That's a significant performance drop. The phone is still comfortably above 3k elo, but not 3500. With prep and meta knowledge in a 100 game match? You only need to trick the machine once.
Okay I’ve never seen that 15 second game strategy before, that’s pretty amazing to literally be able to react faster than stockfishjs. I’ve seen that guy do puzzle rushes before he is an insane talent.
I guess I am overestimating the difficulty because I haven’t seen recent anti-computer strategies. I also know they NNUE changed a lot because contempt is no longer a thing.
I personally find this saying extremely overrated and damaging in many cases. Especially considering the pacifying effect it can have on attempting to undo certain injustices.
Best use of it is in your personal life with family, friends, retail worker in your local grocery shop etc.
I agree. My guess is that institutional malice is fully explained by economic incentives 99% of the time. Which the version from the other comment cover: "Never attribute to stupidity, that which can be adequately explained by economic incentives"
Maybe the AI considered that after joking together for a while it could make a last one, touching a very sensitive subject but at the same time not being offensive at all.
How will be differentiating a malfunctioning AI from one that is too subtle for us? Fun times.
It could also be that they see no scenario ahead where they want the average number of children per woman to be above 3. I don't think China wants back to exponential growth ever. They probably want to find their equilibrium.
Consider if something unlikely happens that completely changes the dynamic of families, e.g. a cultural shift where women stay more at home and have more babies. If the average family started having more than 3 children, they would consider bringing back this system, which could cause big social problems, as it costs more social capital to bring back an unpopular system than to simply maintain it.
I think the idea is to bring the average up to 2.1, which is the accepted "replacement" fertility rate, to stem a population decline. One way to achieve that level is when a majority of women have 2 children but a few have 3 children. For a population the size of China, a "few" might mean a few million. I don't see that happening without a few carrots.
I am not sure, that this is a problem that needs centralized "solving".
I don't want anyone tell me, how much children I can or should have.
We got to make sure the work life balance is in order for people to have the time and money to afford a family and give adequate support like daycare etc.
But apart from that, I really don't like some buerocrats in some buerau somewhere calculating the "correct" number of births. That regulates itself. There is immigration and emigration. There is automatisation (elderly care), better medicine, so older people can tend to themself longer and less need of a "dumb" workforce etc. etc.
Trying to calculate it and declare meassures based that, can only fail in my opinion.
> We got to make sure the work life balance is in order for people to have the time and money to afford a family and give adequate support like daycare etc.
Stuff like this can and does result from bureaucrats / policy-makers deciding what an ideal number of children (for the well-being of the society) is. You say "that regulates itself", but clearly the things you listed (work-life balance, economic prosperity, daycare, etc.) are not self-regulating. You also mention specifically that this matter doesn't need centralized solving, but that's different from what you went on to say, that the factors that go into determining how many children are had are self-regulating. A decentralized approach does nothing to guarantee that.
Whether you like it or not, the government has a large influence over all of these matters. Choosing to not regulate them is a choice that will affect how they turn out. Likewise, the very structure of the economy and social systems will affect these things. A government can not avoid determining what those are like, whether by the government's active influence over them, or its more laissez-faire approach. When you have a monopoly on violence, you cannot truely recuse yourself from what happens in the society around you, and if you lack a monopoly on violence then you aren't really a government.
I have a young family, so I can tell you that I do know a bit how things go and how many things don't work so well - to which I do in fact blame the various regulations.
Because you know, what worked best for us? All the things that are not regulated, like grandparents watching over the childs or teenage babysitter.
The very well regulated state kindergarten?
It was a nightmare so far, even though we have a quite good kindergarden compared to the various stories I heard so far of what is possible, as well. And sure, Corona was not helping with that either, but I know quite some people in social jobs and I listen to their stories since way before corona.
Wow... I can't believe you're advocating for something completely normal, like choosing to decide how many children you have, and how creepy it is that some governments literally dictate to you this most private aspect of you and your spouse's life, and you are being downvoted. The shills must be out in full force today.
Considering countries who have tried to increase their birth rate have failed to even budge the number by 10%, I’m pretty sure the risk of too many births is pretty much non-existent.
This has to be looked at country by country. Unlike Europe, China still has a large rural, poor population. Without regulation, their population will explode, something the chinese government probably wants to avoid.
Instead of regulating their lives like they do with the child policy and making migration illegal maybe they should try improving their lives a little so they don’t need to have a ton of kids or work illegally and without services in order to get by? Who would have thunk a communist country would treat their poor workers so badly?
that is not really "exporting". If you export bananas, the bananas do not come back to you, while these workers are only temporarily abroad, they're not (for the most part) emigrants who moved there to stay.
The 3 children per woman limit is still the smallest number that accommodates for the 2.1 average required for replacement level.
To achieve equilibrium and avoid rapid growth the limit must be set to 3 and then hope that the average will be drawn down enough by the women who can't/won't have children for various reasons.
Not advocating for this policy, and I am not considering how well it works in practice and the morality. This is just a reductionist model.
When X is 2, then 2 people replace 2 in theory, so there is no growth. In practice, some of those children will die early, so 2 is actually a minor reduction, not growth, forget about exponential.
When X is 2, you have one of the most important functions in the family of functions that display exponential growth. It is the one in which the exponent is 0.
> not growth, forget about exponential.
This is really weird phrasing, since you're on much stronger ground saying "not growth" [arguable] than you are "not exponential" [flat wrong]. I just pointed out that defining results in terms of "X children per woman" will always necessarily produce an exponential curve. That's the definition of an exponential curve. If you want to distinguish between "growth" and "decay", you can say so, but you're still stuck with labeling them "exponential growth" and "exponential decay".
> When X is 2, you have one of the most important functions in the family of functions that display exponential growth. It is the one in which the exponent is 0.
This is technically correct, but also purely academic. If I called the function f(X) = 5 in an analysis exam "exponential", then I would be laughed out of this exam, and for a good reason.
> This is really weird phrasing, since you're on much stronger ground saying "not growth" [arguable] than you are "not exponential" [flat wrong].
"not A" implies "not (A and B)". That's all I meant.
> If I called the function f(X) = 5 in an analysis exam "exponential", then I would be laughed out of this exam
That's not true at all; you'd have to look at the context. If you called it an exponential function as part of a discussion of exponential functions, you'd raise no eyebrows.
Even if you were doing it in a weird way, it's unlikely you'd get laughed out of the room; math exams are not known for penalizing you for being correct.
Yes, it is. Since the population is shrinking over time, it is more often called "exponential decay", but mathematically there's no difference. It just means the value of the exponent is negative.
Per this argument, I would think that an exponential function is different to something experiencing exponential growth. One of them is a definition, the other is a description. They aren’t really interchangeable when it comes to communicating a point.
Concurrency/Parallelism is the number one reason I move a project from Python to any other language. It's one the few areas where Python can't provide a gracious frontend or an acceptable runtime, but it is very impressive how much the community is willing to do to mitigate the challenges.
Process/Thread pools help to mitigate the brain damage in Python significantly, but I hear you. Python's concurrency story is the main reason I started looking at other languages for a better concurrency story and it's why I picked up Clojure as a second language.
