OK, let's get started with the chat session. Welcome everyone!

As usual, please keep unrelated discussion until we've finished with the prepared topics

No disagreement! That is why further analysis is needed. But again, images that have consistency with expectations of such phenomena are unlikely -- not impossible but unlikely -- with erroneous data from local phenomena

To start things off, is anyone new to chat, new to chat sessions, or new to the site?

Or any general questions about the site?

Is discussion about site-policy on-topic here?

Heh! Now that it's chat time we all get quiet!!

@Sid yeah, could be, but hold off on that until after our physics time

Let's move on to physics stuff. The big news is, of course, the Nobel prize announcement from earlier today

Welcome me someone I am a newcomer

@samjoe hi ;p

@samjoe you are at wrong place

Welcome newcomer samjoe!

@Fawad That's an excellent welcome. :P

@DavidZ @Terry Hi!

@samjoe welcome 🍪

@Sid Hi!

We were talking about the Nobel research just before the session so we can get back to that, but let's not let that stop us from linking to other recent developments as usual

@Sid oh you could also use the backup room if you don't want to wait

I've been pretty busy myself so I haven't been able to keep up with new science developments the past couple weeks

re LIGO, sorry, typo, was trying to point out/ compare it to the statistical best practices at LHC... Terry have you looked at any papers analyzing LIGO noise? have glanced at some

Only very briefly with that paper that was just listed above. I would like to look at it in more detail. There are some very sophisticated models these days about the interaction between the conceptual model for this the phenomenon, and the data that is coming in. It is always a more Interactive matching process than people tend to think at first glance.

That interaction is in fact a deep and rather fundamental part of cognition in general, one that has become particularly important for astronomical interpretation.

@TerryBollinger one wonders if they were going to award the prize even without the recent Virgo confirmation. but yeah, 3 detectors correlated, bang, nobel prize =D

@vzn Oh yeah, I did see that one briefly. Seems to be quite a startling result, if it holds up.

Good going Kip Thorne

@vzn that's a fascinating topic although I think of it out of scope here.

@vzn The Virgo confirmation was just announced a few days ago, wasn't it? I would think the Academy made their decision before then, although I don't really know

@TerryBollinger how so? Seems on topic to me

unless I'm missing something

You mean the quantum interpretation? Whether we're in a computer? Actually I think it's very relevant to physics,I was just afraid of going off on a tangent from the LIGO discussion.

oh now I want there to be Lego LIGO :-P

Lego LIGO, I love it! Obviously I'm using voice dictation right now.

I'm in a computer right now

@DavidZ have been tracking the "universe as simulation hypothesis" for years, far from an impartial/ neutral observer/ mere spectator on this one, blogged on it extensively, its a strong contrary assertion, but am sharing it in spirit of "full disclosure"

Beep boop

@TerryBollinger If you're actually using dictation software, it's doing really well

Actually, I agree! I've been quite impressed with Google's voice capabilities lately.

@DavidZ that would be my guess also. prize committees involve a lot of "due diligence" which probably spanned many months. but its also rather opaque. but the timing is remarkable/ bordering on huge synchronicity.

Regarding the simulation hypothesis, although I have not yet read that paper I would immediately suggest that there are subtleties to that issue that probably cannot be resolved in that fashion. I would also introduce the concept that there could be such a thing as a naturally occurring computer as opposed to one sitting in a room. That would give a whole different interpretation to this concept.

@TerryBollinger Yeah I'm particularly intrigued that the punctuation is coming through

It's a bit Limited. You can put in the major punctuation points such as periods and commas, but that's about it!

@TerryBollinger agreed. dont really like their conclusions. think the question needs to be subject to experimental testing. the general idea that "calculating the universe state takes gobs of computation, and is therefore infeasible/ impossible" doesnt make basic sense to me. yet there are similar ideas circulating in quantum computation.

I don't write many papers, but this is a topic that I am working on an I think interesting collection of papers centered around the fundamental idea of naturally occurring computer. We'll see how it goes.

top of the morning to you laddies

@TerryBollinger sounds interesting would like to see/ hear more, have probably been delving into similar refs myself, itd be neat if you blogged on it (far less overhead than writing papers)... o_O

@BalarkaSen howdy

Hi @DavidZ

I'll keep some updates here. As usual for me lately, health is a bit of an issue.@vzn

Oh, I see the chat session has started

Yeah, we're halfway through, but it's pretty much just open discussion right now

@TerryBollinger sorry to hear that. how about a sample, what is a major paper nearby to your inquiry?

Or I'll work on researchgate... (Vs blog)

What if we are in a quantum computer simulation?

@DavidZ iirc isnt your new job related to speech recognition? hows it going?

@DavidZ really? Interesting!

If whatever that simulate us is a quantum computer then it can account for any quantum phenomenon. However this argument sounds cheaty

@vzn my primary premise is sufficiently different that I've not found any really close matches in papers that I've seen. Again, the idea that were in a naturally occurring computer in which Quantum phenomena are actually a reflection of computational issues seems to be an interesting spin on the whole idea.

I wonder if it is even possible in principle to rule out a physical phenomenon being simulated by any arbitrary computer like abstract machine

@Secret, I would actually suggest that Quantum phenomena are more a reflection of how the computer works and its limit of resolution. You would be surprised how far you can go with that in terms of understanding why Quantum phenomena exist at all.

@TerryBollinger (spin, lol...) youre in good company, speaking of nobel prizes, 't hooft is working on similar directions. am always amazed/ mystified/ baffled how little attn his work in the area gets...

t'Hooft is a truly amazing man and deep thinker, and I like his papers very much. That's not to say that I understand all of this papers!

@TerryBollinger so, using such framework, you mean quantum phenomenon is something akin to glitches and underflow overflows of the underlying natural computer?

Not so much glitchy as fundamental to how the model works. A partivle only has so many bits of information available to it for its representation, and depending on how you slice those bits up you wind up having large areas of indeterminacy. This makes for a remarkably efficient computstionsl model that I'm also trying to translate into real algorithms. With a little bit of success even!

@TerryBollinger he reminds me in some ways of einstein in his "late life off-the-beaten-path meanderings" and that pointed smbc cartoon... o_O

@TerryBollinger CS rulez =D

@Sid what did you want to ask about site policy?

@vzn Yeah it is - more on the side of understanding than recognition. It's difficult but quite interesting.

@DavidZ that is very cool and very definitely where the whole area is headed. Going to understanding is where we move from simple translation to actual machine cognition. Do you have any papers or anything out on your work?

@DavidZ cool that youre now working close to AI/ CS. what language(s) are you using? soundhound.com

@TerryBollinger Not personally - it's not an academic job so no time for writing papers! Though there might be some papers by the founders of the company from back in the day when it was just a pet project.

@vzn Mostly C++ with some in-house extensions and libraries

Fantastic area in any case, and as someone primarily from computation background I always strongly encouraged physicists to participate! There are so many good things that physicists can add in terms of mathematically modeling these phenomena, especially with your experience in complex math that is completely lacking in the computer science community.

Unrelated: Sometimes, I really felt like earth is passing through a time constriction region of spacetime (whatever that means), cause everything happens so fast and so many in 2017 compared to last year

@TerryBollinger Yeah, although my job is not very mathematical so far, which is a bit of a shame... still I like it

On another note, I am dealing and learning with set theory with the math guys recently, which computational theory is also closely related to

Anyway something's come up and I think I have to run out on the chat session early today (i.e. now)

By David Z good talking to you!

I'll leave you all to close it out (21 minutes left) and see everyone for the next session in two weeks!

@DavidZ great to hear from you

@Secret the interesting thing about set theory is if you analyze it from a computational perspective, or more specifically, from a cognitive perspective. Sets require a contexting mechanism that is closely akin to some kind of cognitive context. This makes set theory considerably more complex than one might think from looking at it just as an abstract set of symbols. There's

re AI/ new science just saw this on reddit, a new theory of hippocampus by deepmind, impressive work from them as always. deepmind.com/blog/hippocampus-predictive-map

... a computational context that can be quite complex for making those symbols work.

@vzn sounds intriguing!

dead silence ... lol what is it about an official mtg that causes all the otherwise noisy room/ regular chatters to back off and turn into lurkers o_O

@vzn I just looked at that site and it has some pretty good stuff on it. Some of the new stuff is closely related to the research I was involved with prior to retiring. That idea of involvement of memory with cognition is extremely important, and something that I feel will be fundamental to going to True machine cognition

@TerryBollinger yeah my latest theory is that AI is actually at core "novelty detection/ seeking" which gives a general framework for understanding memory & lots else... the hippocampus has long been a fascinating study for psychologists via famous patient HM who had his damaged and then lost ability to make new "long term memories"

"Emergent persistence" -- essentially emergent memory at its most minimal levels, but growing much more complex as you go into higher levels such as cognition -- makes a surprisingly interesting crossover theme between fundamental physics and various aspects of information focused systems.

@TerryBollinger emergence one of my favorite words these days. music to my ears. have you written any papers previously?

Trying. Just got out of twelve days in the hospital, again. Interesting ailment I have, I can be going most completely normal and then have this happen.

@TerryBollinger ouch, sorry to hear that. so when you publish something, it will be your 1st?

I would not necessarily advocate emergent universes, but neither am I adverse to them. If you combine the emergent memory theme with fractally variating universes, it's not too hard to come up with emergent memory as a driver for the persistence of certain classes of universes.

@vzn for decades I worked for the federal government, often in areas were technological impartiality was extremely important. One thing that I really like about me retired now is I am free to publish or focus on what I wish to focus on.

@TerryBollinger yeah have wanted to retire myself from young age (for similar reasons) :)

I must be going now, it's been great fun talking with all of you! Although I've not been updating it, you can check out researchgate to see where I'm going to put dates on anything I'm doing.

Bye!

@TerryBollinger nice chatting take it easy

↑ re signal vs noise topic, very nice visualization of LIGO noise levels. notice how similar Virgo signal is to bkg noise, & how superior/ standout the livingston signal is!

@ACuriousMind You here? I have a spinorial doubt

spooky scary skeletons, sends spinors down the $\text{Spin}(n)$

so @JohnRennie I imagine you've got an answer to physics.stackexchange.com/questions/360799/… in the works?

@EmilioPisanty I saw that, but it's the sort of confused question that's hard to answer. I think it would be time consuming to write a really thorough answer. In any case I'm going out drinking in 15 minutes and won't be in a state to answer when I get back :-)

@JohnRennie have an Old Rosie to the health of the expanding universe, then ;-)

@Secret Well deserved, of course, but I question your use of 'finally'. From my point of view this is the earliest it could have been awarded. There simple weren't enough events for an award last year.

@Slereah People have tried to build taxonomies for the words 'law', 'principle', 'theory', 'hypothesis'. All such efforts run into trouble with the less than systematic hisotrical use, of course.

Also who cares really

If you really want to be rigorous you just use "axiom" and "theorem"

But the most common theme for 'law' is a empirical relationship observed to have broad applicability.

As in 'Hooke's law' for springs. Many, many springs are seen to obey that.

@Slereah I can get aboard with that, but I also see the point of people who want those words to mean something.

@Slereah presumably you're also allowed to use "conjecture"?

And if you are reading historical literature it helps to know what (if anything) the authors thought those words meant.

@dmckee Rust!

I'm really confused as to what on earth a colon represents in set theory? Is it just like a | (as in it stands for such that) or does it have some other meaning? An example would be $\mathcal P(A) = \{B : B\subset A\}$

Ahhh thanks! Been looking everywhere for this.

it's the same as |, yeah

So the above example says that it's a set B but B is a subset of A.

it's more than that

it's every such set B such that B is a subset of A

Ahhh okay

so, all subsets of A

Oh cool... This is so weird to get started on... swear I have to have a dictionary of symbols on another tab... should probably learn 'em.

one thing to keep in mind in there is that B is a 'dummy' variable in that it doesn't matter what name you give it; you could have C or x instead

Ahh so actually the above example I used is the mathematical definition of a power set...

right

hence the P :)

Goddamn these symbols

keep on seeing things like $|f(x) - g(x)|$ and keep on thinking it's the absolute value of a function $f(x)$ minus a function $g(x)$.

lol

@CooperCape pretty sure that is what it is

so you're thinking right

Yeah but like set symbols goes "Nah bro it's the cardinality of the relative complement of $f(x)$ and $g(x)$" unless I'm being a complete ninny.

(quite possible)

(very)

idk what context you saw that in

but in analysis stuff or at least the books I read/am reading it is what you said before

Think usually relative complement would be shown as A\B

sighs

But cardinality definitely uses | |

off topic: for some reason I feel really uncomfortable talking about math in this chat because I feel like I will get destroyed immediately by the regulars every time for not being rigorous enough.

what I said only applies when someone asks the chat something (including everyone in the room) and I feel like answering because nobody else did yet.

lol Blue... pretty sure a while ago I notated the distance travelled along a metric as $$s=\int_{t_0}^{t} \sqrt{ds^2}$$ and 0ßelö got a bit salty with me but tbf that's reallly crappy notation lol.

if they're helping you or discussing something with you etc... that's ok.

Yeah I've learnt a lot too... but I'm also insecure af so what can yer do amirite?

(More sets questions sorry) But if I write $x\in A$ does that define $x$ as the whole set? as in all the numbers $x$ could be? Or is that saying $x$ is only one element of $A$?

Ahh okay... so if $x\in A$ and say... $x\in B$ then $A=B$ as they both have exactly the same elements?

ohh I see

So $x$ is really a pick a number range to be defined.

i think you should use diagrams or pictures to learn this beginner set theory set stuff

it's helpful

@0ßelö7 @BernardoMeurer &*%^ fantano's shitposting channel has been terminated

What the actual frick is youtube doing

@BalarkaSen waht? Proof? Reason?

@CooperCape I don't remember getting salty at you

but I'm salty now

youtube.com/user/thatistheplan

looking at his tweets it seems youtube blacklisted his channel

get rekt

@Blue you piece of work

you killed Fantano????

hoooo, boy, this site has some shining examples some times

@BalarkaSen So if I have a sequence $h_n$ of $L^\infty$ functions on a compact manifold such that they are uniformly bounded in $L^\infty$, and $h_n\to h$ in $L^2$, is the convergence in any $L^p$?

It's true for $1\le p\le 2$

dont have the time to think now

got to read some stuff

@BalarkaSen Yeah, I just saw it

No clue what happened

im actually angerey

Link me his twitter

@rob That Davydov-Filippov question might be up your alley?

@BernardoMeurer get back to work.

@DanielSank What work?

on the bright side, idubbbz just dropped a content cop on jake paul

about 4 hours ago

Hmmmm

@BalarkaSen So $h_n\to h$ in $L^2$. But each $h_n$ is $L^\infty$ uniformly. We must have $h\in L^\infty$, right?

yeah

But how to prove that

Maybe it's trivial

A subsequence converges to $h$ pointwise a.e.

Ah, one can then use interpolation theory

(maybe)

Maybe interpolation works for $p=\infty$

maybe I misread the question

@BalarkaSen Morally speaking, if $|h_n(x)|\le\Lambda$ for all $x\in M, n\ge1$, and $L_2\lim h_n=h$, can $h$ blow up?

I mean....

That seems obvious

and up to sets of measure zero that's what $L^\infty$ means

@BalarkaSen I dunno bro, times like this I wish I were an algebraist

Although not an algebraic geometer, that seems pretty hard

heh

@lılostafa ::sigh::

@EmilioPisanty :O

@lılostafa i.e. you're about two years too late ;-)

but nvm

let's not delve into yet another discussion of the topic.

@0ßelö7 Well not salty persay

Lemme find the transcript thing

Just at the time came off salty

maybe I'm remembering incorrectly - sozz

And tbf what I wrote was veeerrrry stupid

OH MY LAWD the idubbbz content cop is the most epic roast i have seen in a while

Ahhh no spoilers pls

lolol ok

I am still dazed

this is too epic for me

much unflattnes

@BalarkaSen WATCHING

!!!!

I started watching with enthusiasm and it doubled after I got to know what its really on

wtf is this

its on the asian jake paul lololol

clickbait trash

@BalarkaSen why did you hype this up

cuz its good, m'dear watson

It was about time someone brought RiceGum down

I have no idea who that is.

wut????

what rock do you live under???

ksi and ricegum are the genre-defining artists on youtube who started the curious content of making diss tracks to other people

@BalarkaSen rock?

a paper on

they are literally like the father of kids like jake paul

piles shit

@BalarkaSen I live under i.gyazo.com/0f34df8de0b65678c1b64029b2c29ffe.jpg

beautiful

@BalarkaSen Sadly I just made that so the books I'm actually using for my talks are at the bottom of the stack

10/10 merchandise

@0ßelö7 Your shelfs are exploding, I presume

@BalarkaSen exploding?

Oh, I misread the message. You made the stack so that the books for your talks at the bottom of the stack, not the shelf

Why??

the books I am currently reading were off to the side to make room for my [redacted] when I reach the climax of the content cop

I pulled books from the shelf and floor to make the pile

it was an artistic tower

lol

(replace the art with aut)

lolol

an automorphism tower

ur clearly doing too much algebra

wait oh my god

OH MY GOD

the video got baleeted

LMAOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

DRAMALERTNATION

I'm only 12 mins in

You can still see the thing?

it's still playing

Huh

It will probably fail to buffer in a bit

Damn man the last 3 minutes were the shit

he actually made a disstrack to ricegum

and every fucken youtuber starred in it

incl Pewdiepie, Ethan Klein, jacksfilms, you name it

the disstrack is still up here

it's still going

if I pause or do anything it will die

I have to pee :(

lolololol

hold it right there

I'm 99% YT can't delete the video

nothing he said is worse than RG's content or disstracks

does it not give you a delete reason

it was because of spam lol

so is RG's content, or the whole fricking channel of idubbbz

I'm pretty sure there's a good appeal that can be done here

you're just a hater

and we $\mathfrak{dab}$ on people like you

lol

he made a song "god church"?

yep...

It was with Alissia Violet I think, JP's ex

I feel like ACM now

I don't know any of these people

everything comes back to JP

that's the real person you should know

@BalarkaSen ey, yo

do u wanna Lp now

song was ok

dunno who the ACM-looking guy was

Fuh sure

I didn't recognize one guy in there

ok so $||h_n||_{\infty}\le \Lambda$ and $h_n\to h$ in $L^2$.

Since $M$ is compact, we have $h_n \to h$ in $L^q,$ for any $q\in [1,2]$.

Now I need $h\in L^\infty$.

are any particles observed with a changing rest mass?

For a subsequence, we have $h_{n_j}\to h$ pointwise.

So $h$ must be essentially bounded since each of those are.

So we have $||h_n-h||_\infty\le 2\Lambda$

@BalarkaSen Good so far?

Ok, I agree

So I'm worried about the convergence for $p\in (2,\infty]$. I don't know if the convergence should work in $L^\infty$, actually.

@BalarkaSen Do you know the generalized Holder inequality?

I know Holder, but not generalized Holder

$||uv||_r\le ||u||_p||v||_q$ if $1/p+1/q=1/r$

I want to combine $L^2$ convergence and $L^\infty$ boundedness to get $L^r$ convergence

Oh OK

such interpolations are common but there's a mess of exponents that I can never remember

I see what you're trying to do, that's pretty smart

@ACuriousMind Halp

Hegel is rekin me

@BalarkaSen in theory I get some term like $\sim\Lambda ||h_n-h||_2$

His Philosophy of Right is confusing when applied to intellectual property

gg, idubz

@BernardoMeurer I don't know anything about Hegel except that suspiciously many different people think they stand in his tradition :P

@ACuriousMind ``Stand in his tradition''?

@BernardoMeurer Uh

"think that their views are derived from his"? Yeah, I think that's what I wanted to say

I just ate, brain isn't working too well :P

it's midnight in Germany

That is correct

@ACuriousMind Ah, I see, yes, I will second that

you should be sleeping, not eating

@0ßelö7 You're not my real mom!

@ACuriousMind Hmm, I'm not so sure

i can't sleep today

gotta keep up with all the drama

Lol

@BernardoMeurer yeah this proof works.

I actually need a different inequality though

What proof?

lol wrn png

@BalarkaSen $1\le p\le q\le r$, $$\frac{1}{q}=\frac{\lambda}{p}+\frac{1-\lambda}{r}\implies ||f||_q\le ||f||_p^{\lambda}||f||_r^{1-\lambda}$$

choose $p=2$ and $r$ very large. Bound the $L^r$ norm with the $L^\infty$ one.

Aha

@BalarkaSen Compactness is actually key here. See math.stackexchange.com/questions/811882/… .

Oh wait. Does convergence in every $L^p$ imply convergence in $L^\infty$?

Oh, I don't think so

It's known that $\lim_{p\to\infty} ||f||_p=||f||_\infty$.

(if $f$ is in every $L^p$)

but that's probably not uniform so you can't switch the limits

hm

I can't come with an example off the top of my head

what if something is like $o(x^{1+\epsilon})$ for all $\epsilon$

like the log. that's L^p for all p aint it

if log is integrable I'm crazy

well what is your domain

full real line or what

compact manifold

log is not L^\infty

ze log is integrable over [0, 1]

It's not going to be highly integrable on that domain.

hmm

I'm trying to find a proof of that norm convergence result

page 11 of CFA

Oh it's back up now

excellante

oh that limit can be infinite

maybe log is highly integrable. I doubt it, squaring a singularity makes it worse.

huh? 1/x

the more you take power the more integrable it is

at $0$?

oh, ok, i misunderstood

i was thinking about 1 to infty

pardon moi

these guys only say "for all $p>1$"

I can't believe that "Lp convergence for every p" doesn't come up with anything on google

analysis is scarce

@BalarkaSen Oh well, good question for my PDE class tomorrow

:P

@BalarkaSen Why the tongue?

:P