mwacksen

Karma: 81

mwacksen 29 Oct 2022 10:49 UTC
19 points
5
on: Ukraine and the Crimea Question

In the Nüremberg trials, we decided that one aspect of our global democratic culture is that those who engaged in the Holocaust and mass murdered Jews and other groups were war criminals even if their excuse was that they followed orders.

While I agree with the conclusion, I really do vehemently object to the framing. Who is “we”, and what on earth is “global democratic culture”?

Attempts to rewrite history and glorify fascists go both against our general Western consensus

I mean these are two separate things, and depending on who you ask rewriting history is what the West is pretty good at. But again, what “Western consensus”? Depending on who you ask, either Russia or the West is massively “rewriting history” regarding Ukraine.

Stepan Bandera was at a time a leader of the faction of the Organisation of Ukrainian Nationalists in Kyiv where Ukrainian nationalists together with German death squads committed pogroms.

If I’m not mistaken, Bandera was interned at a German concentration camp when the atrocities took place. Which is quite a major detail, “our Western consensus” generally considers people innocent by default.

From my German perspective, where we consider the Holocaust as one of the worst episodes of the 20th century, declaring people who participated in the Holocaust to be national heroes who shall not be criticized feels deeply wrong

Ah, a German perspective, this somehow fits in very well with the extreme focus on nazis and/or nazi symbols as a kind of axiom from which everything else is derived. Here’s an anecdote: I was friends with an Ukranian, who once wore a stylized swaztika necklace. Was she a Nazi? She was adamant that she was not. Turns out she just liked the pattern, had seen it being used in India, and didn’t treat it as anathema the way a German would. The take that symbols that were used by nazis aren’t proof of evil is an underrated one (especially in Germany). It’s your actions that define you, not which symbols or Russell conjugation you use (or don’t use) when talking about them (“I am a patriot, he is a nationalist, they are a government with a democratic mandate to put their population first”).

That said: nationalism really is a problem in Ukraine at the moment. But I’m more worried about the kind of nationalism the West tends to support (making sure everyone says “Kharkiv” instead of “Kharkov”, calling Russian soldiers “orcs”, insisting that Crimea “is” Ukranian), rather than the “using Nazi symbols sometimes” form.

Right-wing ethnonationalism seems to me the best explanation for Ukrainians needlessly creating both internal conflicts that alienate internal minorities and conflicts with Europeans, at a time when Ukraine wants European support.

Unfortunately I don’t think this harms their European support much—these are all things that are far in the past, and haven’t stopped the EU waxing poetic about how Ukraine is fundamentally one of them. In fact, since in Ukraine, nationalism = against Russia, and “against Russia” is popular in Eastern Europe (for obvious reasons), the nationalist aspects may even find them support from the West. That said, it’s just morally a bad policy, and absolutely affects how Ukraine is seen by Russia.

If we believe that Russia might have manipulated the counting of the referendum, the straightforward way to get a sense of what the Crimeans think is to look at polling data.

Unfortunately not, people who don’t want to be in Crimea will have left, and polls aren’t always reliable.

When pollsters asked again in 2017 in Crimea, 86% of non-Crimean Tatars say they would expect the same result in a repeat referendum

The non-Crimeans Tatars are somewhat irrelevant here given that most are Russian.

Punishing Crimeans for voting for the annexation feels morally wrong to me

There’s a strong case that punishing Syrians for being ruled by Assad might just feel morally wrong too, but who am I to argue with a moral authority like the “Western consensus”.

The fight of Ukraine against Russia is not one with clear lines between good and evil

Thank you for the great post on the clusterf*ck that is Ukraine and Ukrainian politics. Really this is a great case of a situation where it isn’t clear what is actually going on given the rather one-sided reporting on both sides.

Crimeans deserve [..] a peace agreement [...] that respects the referendum of 2014

No they don’t. The “referendum” of 2014 was a joke. Nobody needs to respect this joke.

mwacksen 5 Nov 2020 11:56 UTC
14 points
on: Multiple Worlds, One Universal Wave Function
I’m very confused by the mathematical setup. Probably it’s because I’m a mathematician and not a physicist, so I don’t see things that would be clear for a physicists. My knowledge of quantum mechanics is very very basic, but nonzero. Here’s how I rewrote the setup part of your paper as I was going along, I hope I got everything right.

You have a system $S$ which is some (seperable, complex, etc..) Hilbert space. You also have an observer system O (which is also a Hilbert space). Elements of various Hilbert spaces are called “states”. Then you have the joint system $S \otimes O$ of which $Ψ$ is an element of, which comes with a (unitary) time-evolution $E : S \otimes O \to S \otimes O$ . Now if $S$ were not being observed, it would evolve by some (unitary) time-evolution $E_{S} : S \to S$ . We assume (though I think functional analysis gives this to use for free) that $(v_{i})_{i}$ is an orthonormal basis of eigenfunctions of $E_{S}$ , with eigenvalues $(λ_{i})_{i}$ .

Ok, now comes the trick: we assume that observation doesn’t change the system, i.e. that the $S$ -component of $E$ is $E_{S}$ . Wait, that doesn’t make sense! $E$ doesn’t have an “ $S$ -component”, something like an $S$ -component makes sense only for pure states, if you have mixed states then the idea breaks down. Ok, so we assume that $E$ , when acting on pure states, is equal to $E_{S}$ . So this would give $E : φ \otimes ψ \mapsto (E_{S} φ) \otimes ψ_{φ}$ , where $ψ_{φ}$ is defined so that this holds. Presumably something goes wrong if we do this, so we instead require the weaker $E : φ_{i} \otimes ψ \mapsto (E_{S} φ_{i}) \otimes ψ_{i}$ . And bingo! Since the $φ_{i}$ are eigenfunctions, we get that $E (φ_{i} \otimes ψ) = φ_{i} \otimes (λ_{i} ψ_{i})$ , and let’s redefine $ψ_{i}$ to include the $λ_{i}$ term because why not. Now, if we extend by linearity we get that $E : ϕ \otimes ψ \mapsto \sum_{i} a_{i} φ_{i} \otimes ψ_{i}$ . Applying $E$ again gives $\sum_{i} a_{i} φ_{i} \otimes (ψ_{i})_{i}$ , and the same for further powers.

Ok, let’s interpret that last part in terms of “observations”. If we take states of the combined system $S \otimes E$ , then time-evolution maps pure states with only a $v_{i}$ component to pure states with only a $v_{i}$ component. Wait, that’s exactly what we assumed, why should we be surprised? Well yeah, but if you started out with some linear combination of eigenfunctions, these will be mapped to a linear combination of pure states, and each pure state in this linear combination evolves as assumed, which may or may not be abig deal to you. In a mixed state that is a linear combination of pure states, we call each pure state a “separate observer” or something like this. Of course, mixed states in a tensor product state cannot be uniquely be written as a sum of pure states. However, if we take our preferred basis $(v_{i})_{i}$ and express our mixed states as pure states with respect to that basis in the $S$ -component, this again makes sense.

So it’s super important that we have already distinguished the eigenfunctions of $E_{S}$ at the start, we unfortunately don’t get them out “naturally”. But I guess we learn something about consistency, in the sense that “if eigenfunctions are important, then eigenfunctions are important”.

Ok, now assume our system $S$ is itself a tensor-product of $N$ subsystems $S_{1} \otimes \dots \otimes S_{N}$ , which we think of as “repeating a measurement”. Now what we get if we start with some pure-state is (in general) a mixed state which can be written as a linear combination of pure states of eigenfunctions. As the eigenfunctions of the different systems are different (they are elements of different spaces), if you start out with some non-eigenfunction in each subsystem, you’ll end up with some mixed state that contains different eigenfunctions for the different systems. The “derivation of the Born rule” doesn’t need this step with multiple systems. Basically, we can see this already with just one system. If we start with a non-eigenfunction $\sum_{i} a_{i} φ_{i}$ , then this gets mapped to some linear combination of pure states via the time-evolution. As the time-evolution is unitary, and the |a_i|^2 sum to 1, we can see that each pure state has “length” |a_i|^2.

Thanks for the great paper! I think I’ve finally understood the Everett interpretation.I think the basic point is that if you start by distinguishing your eigenfunctions, then you naturally get out distinguished eigenfunctions. Which is kind of disappointing, because the fact that eigenfunctions are so important is what I find weirdest about QM. I mean I could accept that the Schrödinger equation gives the evolution of the wave-function, but why care about its eigenfunctions so much?

Covid-19 IFR from a Bayesian point of view

mwacksen4 Nov 2020 23:55 UTC

12 points

2 comments3 min readLW link

mwacksen 5 Mar 2023 21:55 UTC
7 points
3
on: The Waluigi Effect (mega-post)
I understand that—with some caveats—a waluigi->luigi transition may have low probability in natural language text. However, there’s no reason to think this has to be the case for RLHF text.

mwacksen 30 Oct 2022 8:26 UTC
7 points
−3
in reply to: throwaway62654’s comment on: Ukraine and the Crimea Question

I don’t see better options. What would you suggest?

The situation is pretty difficult,

a) Providing clear, unambiguous and automatic mechanisms for repealing Western sanctions that are also realistic (i.e. not “we drop sanctions once you get out of Crimea”). While I also like Bryan Caplan’s suggestion to offer asylum + military compensation to defecting Russian soldiers, I recognize that this is politically not feasible.

Sanctions that are not directed at the military antagonize the Russian population. Somewhat counterintuitively, they can even lead ordinary Russians to increase their support for Putin.

b) Recognition of the annexation of Crimea, subject to a (future) internationally recognized referendum legitimized by a UNSC resolution and potentially some kind of autonomy within Russia. Ideally, this would be modeled on Kosovo or Hong-Kong in some way.

c) Some form of Minsk III that gives hard guarantees to Donetsk/Luhansk for a referendum, or makes them de facto independent states that are de-jure part of Ukraine.

d) Making weapons-deliveries to Ukraine conditional on Ukraine stating that it might be willing to give up Crimea + DNR/LNR under suitable terms. Something like this for the Russian side could be (a).

While I recognize that Russia will try to game this as far as possible, the main issue I see with this is that it may incentivize similar behavior by other states in the future. Borders are supposed to be fixed not because this is “fair” or anything, but to avoid people starting wars over borders.

mwacksen 30 Dec 2021 14:39 UTC
7 points
in reply to: CuriousApe’s comment on: COVID Skepticism Isn’t About Science
Funnily enough, I have yet to read a single not-completely-ridiculous cost-benefit analysis that goes this way. We must live in different bubbles.

mwacksen 29 Oct 2022 11:11 UTC
6 points
−2
in reply to: mruwnik’s comment on: Ukraine and the Crimea Question

But so far only one side has rape as part of its doctrine.

If you think Russian official military doctrine includes rape, then (and I’m trying to put this as politely as possible) you are deluded.

Only one side is engaged in wide spread plundering etc

Probably true, but I do wish we’d actually know how much plundering is going on relative to how many soldiers are there.

That seems like an important distinction.

I saw some pretty nasty psychological warfare-type stuff the Ukranians were doing that the Russian’s weren’t. Like sending pictures of their dead sons to Russian mothers, and it seems like Ukraine is potentially willing to kill ‘pure’ civilians abroad than Russia is (see the murder of Darya Dugina, yes I know about Russia going after ex-spies and arms-dealers, so it’s not so clear cut). Obviously it’s hard to compare two countries when one is invading, and the other is defending. But nobody can claim that either side hasn’t committed atrocities.

mwacksen 30 Oct 2022 7:50 UTC
5 points
1
in reply to: ChristianKl’s comment on: Ukraine and the Crimea Question

Bandera was in prison when the atrocities in 1943 took place but not when those in 1941 took place.

First of all, thanks for catching this, I was mistaken. That said, it seems somewhat more complex, according to this link “Bandera was in occupied Poland when on June 30, 1941, his comrades proclaimed an independent Ukrainian state in Nazi-occupied Lviv — and the Germans banned him from traveling to Ukraine.”

This doesn’t, of course, vindicate him in any way—he was head of an organisation that performed atrocities and worked with Nazi Germany. But it also doesn’t make him guilty of said atrocities.

It was also the case before 2014 that the majority in Crimea was Russian and there were a lot fewer Crimean Tatars.

I’m loosely familiar with the history of Crimea, my point is that non-Crimean Tartars (many of who live, and have always lived in Russia) are irrelevant to Crimea. But maybe I’m misunderstanding things, and you mean “non-Crimean” Tartars who happen to live in Crimea?

But even if you want to ignore the Russians the poll also separates out self-identfying Ukrainans and a majority of them were also in support of the referendum.

So what? It’s a feature, not a bug, of the modern system of states that not everybody who wants a referendum to secede gets one.

It was a reason why Ukraine got less military support from the EU before the invasion in 2022 than it wanted.

The reason Ukraine got less military support from the EU than it wanted was primarily so as not to antogonize Russia, as far as I can tell.

There are probably also a lot of others who at the European institutional level think “We already have enough problems with Polish and Hungarian nationalists, do we really want to deal with Ukrainian nationalists as well?”

Yes, hence Scholz’s ridiculous pivot away from unanimity in the EU in Prague recently.

mwacksen 10 Dec 2021 20:13 UTC
5 points
on: COVID and the holidays

An average vaccinated 30yo now loses about 6 weeks of expected life from contracting COVID instead of 2 weeks, because of waning vaccine efficacy.

The counterfactual here is “you never get covid”, so I’d take this number with a large grain of salt. If, on the other hand, the counterfactual is “you get covid a few years later”, then the loss of expected life does not occur. Additionally, if you do get covid, you’re (probably) super immune for a while, which presumably increases your quality of life.

mwacksen 7 May 2022 23:28 UTC
4 points
on: Hard evidence that mild COVID cases frequently reduce intelligence
Correlation is not causation. If the “best” data is garbage, it’s still garbage. We should not update our priors based on garbage data.

mwacksen 30 Dec 2021 14:50 UTC
4 points
in reply to: gabrielrecc’s comment on: COVID Skepticism Isn’t About Science

Not exactly a fair description of what the public health measures have been. What country has been in lockdown for “a year or two” (besides China)?

Is there an end in sight? What country hasn’t had covid-related public health measures—especially on entry and exit of the country—for the last year? It may not have been a full lockdown (whatever that means) for most of the time, but public health measures have certainly been in place.

mwacksen 30 Oct 2022 17:07 UTC
3 points
0
in reply to: ChristianKl’s comment on: Ukraine and the Crimea Question

I did refer to people who identify as either Ukrainian or Russian and not as Crimean Tartar when I said non-Crimean Tartar. So non-‘Crimean Tartar’.

Aaah got it. My bad.

mwacksen 30 Oct 2022 7:41 UTC
3 points
0
in reply to: Viliam’s comment on: Ukraine and the Crimea Question
It seems somewhat more complex, according to this link “Bandera was in occupied Poland when on June 30, 1941, his comrades proclaimed an independent Ukrainian state in Nazi-occupied Lviv — and the Germans banned him from traveling to Ukraine.”

mwacksen 29 Oct 2022 11:31 UTC
3 points
0
on: Translated transcripts of Putin’s major speeches from Oct 27, 2022

Bear in mind that foreign authoritarian presidents are generally not considered reliable sources of information

I don’t think the “foreign authoritarian” qualifier is necessary here...

mwacksen 29 Oct 2022 10:58 UTC
3 points
3
in reply to: throwaway62654’s comment on: Ukraine and the Crimea Question

For some reason public discourse in the Western countries gravitates towards either “let’s stop helping Ukraine” with weak justifications like “will of the Crimean people” and “Ukrainians aren’t saints too” or going all in up to directly fighting Russian army on the ground.

Yes, this is a good point.

I think policy of helping Ukraine but not engaging Russian army directly (basically just sticking to what’s already being done) is superior to either extreme.

Wait what? I mean yes, obviously this is superior to batshit crazy options like sending troops to Ukraine or telling Russia it owns Ukraine now, but in terms of nuclear risk and global security it’s pretty stupid. The current approach incentivises both Ukraine and Russia to continue and escalate as they both ultimately see themselves as prevailing given enough time.

mwacksen 30 Dec 2021 14:43 UTC
3 points
in reply to: Sune’s comment on: COVID Skepticism Isn’t About Science
Population density is entirely the wrong metric to look at here. You could fudge the denmark “population density” count by just including Greenland, and including the empty swathes of land in the nordic countries has the same effect.

mwacksen 6 Nov 2020 10:07 UTC
3 points
in reply to: SymplecticMan’s comment on: Multiple Worlds, One Universal Wave Function
I think my question isn’t really well-defined. I guess it’s more along the lines of “is there some ‘natural seeming’ reasoning procedure that gets me QM ”.

And it’s even less well-defined as I have no clear understanding of what QM is, as all my attempts to learn it eventually run into problems where something just doesn’t make sense—not because I can’t follow the math, but because I can’t follow the interpretation.

If we accept that mutually exclusive states are represented by orthogonal vectors, and we want to distinguish mutually exclusive states of some interesting subsystem, then what’s unreasonable with defining a “measurement” as something that correlates our apparatus with the orthogonal states of the interesting subsystem, or at least as an ideal form of a measurement?

Yes, this makes sense, though “mutually exclusive state are represented by orthogonal vectors” is still really weird. I kind of get why Hermitian operators here makes sense, but then we apply the measurement and the system collapses to one of its eigenfunctions. Why?

mwacksen 5 Nov 2020 9:29 UTC
3 points
in reply to: Linch’s comment on: Covid-19 IFR from a Bayesian point of view
Thanks for the comments!

I agree with all of these points. In fact, with respect to (4) it is even plausible that some “once infected” people never go on to develop the kind of antibodies that are being tested for. Point (3) is why I control for age/sex, but of course there are a number of further complexities.

These further complexities, along with (1) and (2) are currently “un-modelable complexities” for me. There are just so many selection effects in play that it isn’t clear if you gain anything from trying to take them into account. Given that there are a number of papers that try to calculate some kind of IFR making a number of basic mistakes, I wanted to set out to see what the data + simple models gets if you do the maths correctly, as opposed to doing ridiculous things like caring about the median study like the Ioannadis meta-analysis does. After all, it seems like the IFR is what everyone cares about, so it would be nice if we were doing things “less wrong” here.

mwacksen 6 Nov 2022 22:31 UTC
LW: 2 AF: 1
1
AF
in reply to: michaelcohen’s comment on: AI X-risk >35% mostly based on a recent peer-reviewed argument

Care to bet on the results of a survey of academic computer scientists? If the stakes are high enough, I could try to make it happen.

No, no more than I would bet on a survey of <insert religious group here> whether they think <religious group> is more virtuous than <non-religious group>. Academics may claim that peer review is to check validity but their actions tell a different story. This is especially true in “hard” fields like mathematics where reviewers may even struggle to follow an argument, let alone check its validity. Given that most papers are never read by others, this is really not a big deal though.

But I’ll offer three further arguments for why I don’t think peer review ensures validity.

Argument 1: a) Humans (including reviewers) make mistakes all the time, but b) Retractions/corrections in papers are very rare.

Unless academics are better at spotting mistakes immediately when reviewing than everyone else (they are not), we should expect lots of peer-reviewed articles to therefore have mistakes because invalid papers rarely get retracted.

Argument 2: Computer science papers don’t always include reproducible software, but checking code would absolutely be required to check validity.

Argument 3: It is customary to submit papers that are rejected by one journal to another journal. This means that articles that fail “peer review” at one journal can obtain “peer review” at a different journal.

PS: For CS it’s harder to check “validity”, but here’s how papers replicate in other fields: https://fantasticanachronism.com/2021/11/18/how-i-made-10k-predicting-which-papers-will-replicate/

mwacksen 31 Oct 2022 13:58 UTC
2 points
1
in reply to: throwaway62654’s comment on: Ukraine and the Crimea Question

The goal of sanctions is not to incentivize regime change. The goal is to make it a bit more difficult for him to wage the war.

A nice story, but I don’t buy it. How exactly does banning Russian flights to Europe, or banning Russian tourists, or banning Russian bank accounts with more than €100k from transacting make it a bit more difficult for him to wage the war? Or confiscating/stealing the wealth of oligarchs? If Putin “doesn’t care that much” about sanctions, isn’t it pretty stupid that the West is shooting themselves in the foot, and the developing world in the face, by still applying heavy sanctions?

(b) Putin will never allow an election or a referendum that he doesn’t control.

Maybe not, but better than not trying at all. Putin’s rhetoric about the West being hypocritical is spot on, except for the fact that Russia is just as hypocritical. Offering Putin the “rules-based” and law-based international order he has been asking for since at least 2008 is in my opinion a no-brainer, even if he goes on to reject it.

(c) Russian constitution now says Donetsk and Luhansk are parts of Russia.

Yes. Though I’m not sure it’s the constitution that says this or the law. I don’t see how this problem is unique to what I’m proposing, if Ukraine just gradually keeps winning like it is doing right now, the Russian constitution/law will still say this, and it will be an even bigger problem.

The problem is it doesn’t solve the fundamental problem. Which is how Ukraine will protect itself against the next much bigger invasion.

Again: I see this argument thrown around a lot, but nothing solves this fundamental problem, so the existence of this problem proves nothing. Even a long dragged out war doesn’t stop Russia from showing up again, with more firepower, in the future.

Fundamentally, Russia has nukes and Ukraine does not. In this sense, anything that de-escalates this war, while at the same time convincing Russia and China that trying something like this is a bad idea, is the way to go. If Russia decides that it doesn’t like the way the West is supporting Ukraine, it has the means to escalate. In the end, the way I expect this to go is for Russia to credibly threaten/use some kind of nuke in Ukraine, and thereby forcing an agreement with the West. Alternatively, the West gets bored of Ukraine after another year of fighting and Russia slowly but surely takes the regions it has annexed. But saying “I hope everything keeps going like it is going” is magical thinking—the war will end, and I think it makes more sense to reason about the end-state than to pretend it can keep going like this forever.