« first day (3096 days earlier)      last day (516 days later) » 

12:30 AM
@enumaris too bad. what is the direction? btw theres also a closely related area called "sparse rewards" also getting a lot of attn lately but yeah it all tends to be more in the unsupervised area which is maybe less connected to data science/ datamining focusing more on supervised algorithms. (actually cant easily think of unsupervised applications in those areas except for maybe classification...)
@danielunderwood the thing that requires least modification is to actually execute on my existing plan
I think it's good enough to execute on
apparently my manager thinks otherwise
12:54 AM
The wonderful world of corporate decision-making
1:26 AM
2 hours later…
3:24 AM
@enumaris geez sure is quiet around here... wondering wheres your avengers luv this time around? ;) Game over! Avengers: Endgame breaks box-office records as it grosses $140 million in just ONE day smashing the previous record set by Star Wars: The Force Awakens... as fans line-up to watch the film until 4.30am dailymail.co.uk/news/article-6963997/…
Just for fun: Researchers embarked on a novel study intent on measuring what a Princeton philosophy professor contends is one of the most salient features of our culture — the ability to play the expert without being one.

Or, as the social scientists put it, to BS. https://www.washingtonpost.com/business/2019/04/26/rich-guys-are-most-likely-have-no-idea-what-theyre-talking-about-study-finds/?utm_term=.c6a488478974
> “proper numbers,” “subjunctive scaling” and “declarative fractions”
lol! trivial + mastered all that in kindergarten :P
3 hours later…
6:12 AM
@vzn I go way further than that: I have a history of converting nonsense into sense
for example, this engineering joke:
in The Skunk Works, Mar 26 '15 at 0:36, by infinitesimal simplicio
A mathematician, a physicist and a engineer are asked by a student what the meaning of $$\int \frac{1}{dx}$$ is.

The mathematician says it is meaningless.

The physicist ponders it for a moment and wonders if there is some way to give it meaning.

The engineer says, "Hmmmm, I used to know how to do that."
in Mathematics, May 19 '18 at 8:25, by Secret
$$\int \frac{1}{dx} = \int \frac{1}{1-(1-dx)} = \int I + (I-D)x + (I-D)^2 x + (I-D)^3x + \cdots = \int () + \int x() + x() + \int x() - \int Dx () - \int xD() + Dx () + \cdots$$
Having said that, in order to formally create the domain of "proper numbers", we need to know what notion of "proper" means. It is clear that "proper" cannot be size related, since there isn't much room we can extend when talking about size of natural numbers
"Subjunctive scaling" is even harder, for "subjunctive" means whatever this domain will become, has to carry a notion of "degrees of belief" We already have bayesian probability to do that kind of job for us, and it is not clear what it means to scale a probability
These researches about bullshit does make me wonder, whether there exists a maximally bullshit term such that it resists any attempt to make sense of it
That is, given a language L where each word has a semantic meaning, is it possible to find some string s such that s cannot be ascribed semantic meaning for any extensions L' of L
3 hours later…
8:54 AM
Though the top n entries are all Scottish engineers and nothing to do with me.
@JohnRennie I've observed that too; I believe it's because Math SE is a much more active site...so Google crawls it much more regularly and the optimization for Math SE pages is better.
@Blue Math Overflow not the Math SE!
Oh, hah. That's interesting...
I've posted once on MO then deleted that post under a hail of withering scorn :-)
> Google can't index the whole web, nor would they want to. They just want to index pages that have a strong likelihood of ranking so they can build the best possible search engine.
From here. I guess it comes down to how the sitemap is implemented. Or maybe Google optimizes for Math Overflow as it, in general, has higher quality content that other SEs.
@JohnRennie I see there's a cricketer and science journalist too. :)
@JohnRennie Lol...happened to me once. :P
9:01 AM
There are a surprising number of John Rennie's. Both the surname Rennie and forename John are quite common in Scotland. Originally it was a Gaelic name.
Indeed; I remember we discussed it once. :D
When I try searching "John Rennie" from an incognito tab, the Math Overflow page does not appear in the search results...which is weird. Perhaps there's a location-based component...(that is a screenshot from page 2 of the search results, by the way).
@Blue ah yes, I get the same. How odd.
9:20 AM
Where's the mystery? when you're not incognito, Google knows you've been to SE a lot and are likely more interested in hits from SE sites than the average user? (Also, I'm moving this discussion to the h bar since it doesn't really have anything to do with the election)
17 messages moved from 2019 Physics Moderator Election Chat
@ACuriousMind The fact that it optimizes for MO although JR visits PSE way more often.
Which probably implies that MO is somehow ranked higher than the rest of SE, by Google...which is the weird/interesting part.
Probably an effect of MO not sharing the stackexchange.com domain
Although that would also mean SF and SU should appear there
2 hours later…
11:43 AM
Q: Differentiating Propagator, Greens function, Correlation function, etc

Nikolaj-KFor the following quantities respectively, could someone write down the common definitions, their meaning, the field of study in which one would typically find these under their actual name, and most foremost the associated abuse of language as well as difference and correlation (no pun intended)...

I'm curious whether terminology "Green kernel" is used commonly in the same context as this answer.
1 hour later…
wow wow
3 hours later…
4:14 PM
Emergent phenomenon A: Cannot exist if the source B is removed. Thus one must find A and B together in some form
Actual matter A: can exist independently of some matter B
We can thus bump this up to metaphysics and define something as "exists" as follows:
Given concepts, if a concept exists, then it can be separated from another
Note the reverse implication is not necessary true
In order to make that an iff, we need to be more precise
Q: What actual physical chaotic system is most chaotic?

Pedro MedianoCompletely out of curiosity: What actual physical chaotic system is most chaotic, in the sense of rapid error growth? Something that shows (to a non-expert audience) how hard it is to make predictions in a chaotic system. For example, I was thinking of something along these lines: If you sim...

Eh those two answers are next door to NAA
It's a pretty vague question, and the edits made the two answers nonsensical.
Neither answerer is trying to say that their mentioned system is the 'most chaotic' system by any means.
Actually, I'm not even sure how you can characterize a system as "most chaotic": while there are certainly ways to say that certain systems are not chaotic or only slightly nonlinear, I can't think of any reason why there's a point beyond which you can't get more chaotic.
4:29 PM
But the suggestion to make the question "What actual physical system is most chaotic?" was given in a comment by Anders Sandberg, and from his answers, it looks like he knows a lot about nonlinear dynamics and chaotic systems.
hmm, are there perpetual machines of the 0th kind?
or rather, what does a perpetual machine violating only the 3rd law but not the 1st or 2nd law look like
@Chair Well, if there's an answer to that, Quora certainly doesn't have it.
4:48 PM
and then in 2002 there's this weird paper which claimed one can radiative cool something to absolute zero
@Chair It's just not a well-defined question, don't stress over it :P
@ACuriousMind But that's the thing! I can't be sure if it isn't well-defined, because it was suggested by someone who is (or at least seems to be) quite knowledgeable about the field.
Whatever. I'll look at it in a couple of days and see if anything interesting gets posted there.
@Chair His comment says he's asking this in the context of an outreach talk, which makes sense - he's just looking for intuitively "very chaotic" systems, not some well-defined metric.
@ACuriousMind I'd think that if you characterize something as "the most <adjective>", you should have some metric. But perhaps if it's an outreach talk, any example of a very chaotic system would answer the question.
But then that makes it a bit trivial since there're so many well-explained and fascinating examples littered over the scores of pop-sci books.
ok, so basically violation of the third law will mean one can cool faster than expected, thus it controls the rate of cooling
1 hour later…
6:19 PM
What generates the electric field in these universes\spaces: Bianchi, de Sitter?
de Sitter space doesn't have any electric field associated with it
It's a lambda vacuum solution
Bianchi's a pretty large class so I can't speak for that one
I am reading articles about the creation of particles in these spaces with the presence of electric field
6:37 PM
Semiclassical, vzn, Boteppa etc.: Ok after nearly a week of work, I think I started to have some idea on this "quantum theory cannot consistently applied to itself" paper
Don't try to read this pics. I upload it because the working itself looks cool in an artistic sense
Basically, my suspicion is that Assumption C must fail for any interpretations that does not allow a view of the whole universe:
Specially, all the observations in Table 3 applies except for Statement 00:28
The way they calculate that probability $\text{Pr}(ok = \bar{ok}) = \text{Pr}(r=tails)\text{Pr}(z=\lvert ok\rangle) = \frac{1}{3} \frac{1}{2} = \frac{1}{12}$.
Means each Agent observed and had knowledge of the things covered in Table 3 in the paper
However, statement 28 has to be wrong because after $\bar{W}$ does the measurement to the lab $\bar{L}$, there is one in three times it is then projected into the superposition $\lvert \bar{ok}\rangle$. This means while due to born rule in Assumption Q, $\bar{W}$ will be very certain that $F$ will get $z = +\frac{1}{2}$. However, in both $W$ and $\bar{W}$'s perspective, both the labs $L,\bar{L}$ are still in superposition where $r$ is headtails, since the measurement by $F$ is local to $L$.
What paper
Haven't paid attention to this stuff, and seen that this and this claim it has big mistakes
6:52 PM
(cont.) thus it cannot collapse the superposition in $\bar{L}$. Thus to both $W$ and $\bar{W}$, since they are outside of both labs $L, \bar{L}$, thus lab $L$ has 1/2 chance where z is in superposition given by $\lvert ok\rangle$ and thus allow $W$ to get the result in contradiction to what all other agents would deduce.

Ok I am checking those two links now
This one is somewhat similar to my conclusion, that the measurement destroys the information, thus we cannot said "if" for anything that is going to happen in the future
This basically did the extra hardwork that spelled out the state of the Lab $L$, I got the same conclusion semi qualitatively above
7:22 PM
Very similar, especially near the end of the paper when it talks about ultimate observers. But mine is something like, at least for the deductions of F and bar F, they do conclude W will get fail, but because only the labs but not the wigners are entangled, W has 1/2 of the chance to get ok after bar W made the measurement and announced bar ok to W
So my prediction is that, after the experiment when they share the results together, they will indeed find W's outcomes contradict what bar W bar F and F deduced, but that is valid because F and bar F made the reasoning inside the labs (thus they are part of the entanglement) while W and bar W are outside, hence seeing the labs in an entangled superposition
Thus in agreement to the above three links I have read so far, the only thing that W cannot say is he will get fail (statement W00:28 in the original paper), because the labs are still in an entangled superposition before he made the measurement
7:49 PM
I think this paper goes further than mine (and pointed out I am wrong when I say that bar F and F can said anything about W). Basically, given times a<b<c, a measurement at t=b will change what the agents doing experiment at that time can say about other agents in the past and future. He also shared the similar observation on how bar W does not collapse the entangled superposition of the labs, but he goes on a lot further saying how the whole state assignment will be changed if bar W did
not do a projective measurement to the lab bar L
I have not read the more extreme generalisation, but intuitively it would make sense that the time ordering will be scrambled when boosted to different frames of references
Pretty much the above, one observer can assign a superposition, whereas the other can see a definite outcome. Also that one should not perform inference on propositions when a measurement had not occured
8:14 PM
That's a good summary of the above papers
This stuff is like an alternative universe at times, completely irrelevant yet so much energy into it
But it is cool. One reason I am so interested in it is because I want diatheia (physical entities that are true contradictions) to exist, as well I love the weirdness of quantum and how it revise our thinking
While all of the above (plus my own independent deduction) have now showed that the contradiction does not exists, I gained a deeper understanding on how to reason in a a quantum way
Finally this experimental realisation pretty much showed how the labs are not independent and hence an entanglement exists
Thus, I am actually pretty fine with any of the explanations:
1. Objectivity only exists at the global level for some privillaged observer (which is virtually impossible to prove experimentally)
2. Nothing can be deduced about future outcomes based on present and past outcomes. In fact, for spacelike separated events, nothing can be deduced about past and future outcomes based on measurement outcomes at the present
3. It is perfectly valid to have observers getting a definite outcome, while at the same time other observers assigned the observer's system as a superposition, and deduce different probabilities of the outcomes
4. That measurements changes the state means transitivity of implication can fail
2,3,4 are the ones I often used when analysing quantum experiments (by writing out all the states and measurements as rays annotated with observables, and hence tracking how they became entangled or decohered through the process wrt different observers). Acuriousmind said mine is not an interpretation, but quantum mechanics by itself
1 is used when I talked about the collapse interpretations and Bohmian mechanics
In theory, I think if 1 is true (at least for a large enough region), it will be interesting because it means one can think about "controlling predestination" by calculating how an experiment should be set up to get a desired wavefunction
exploiting the fact that the system might have became entangled with the environment and if that contribution is stable enough, then suitable placement of the devices in the experiment can shape the wavefunction into some desirable form
2 hours later…
10:13 PM
Let me say what I am still missing. I read Bier in original and in annex C, it is by no means "obvious". That's why I asked my question in the first place. Next I read your response. It repeats Bier's arguments but I do not see the explicit, new argument either. I would also like to see sb cross over from statistics into physics. May be this is where the Li/s failed. I do still hope, somebody will answer my question. ** "Frankly", you should not have added the emotional paragraph before last. Such doesn't help the community. ** — Juergen 7 hours ago
Does this type of comment tip the question over into non-mainstream?
@EmilioPisanty I don't see why it would. The intent of the asker is irrelevant, what matters is the actual text of the question.
I would have closed it as peer-review/too broad from the beginning, but no one else felt that way, so I didn't.
10:30 PM
@EmilioPisanty Now I'm wondering whether "The Born rule is incorrect" would make an equally good movie premise as the flat earth does
11:01 PM
@ACuriousMind see also OP's deleted previous version of that comment, threatening to flag based on them having taken offence from legitimate scientific criticism. It's not far from requiring moderator intervention.
Though maybe @Qmechanic is better placed to evaluate the technical side.

« first day (3096 days earlier)      last day (516 days later) »