The h Bar - 2018-09-20 (page 2 of 2)

« first day (2877 days earlier) ← previous day next day → last day (2054 days later) »

5:00 PM

photons interacting with neutrinos gravitationally in black holes in an expanding universe

I will note, though, that when it comes to QFT there's basically two ways you can construct a Bohmian story: one with a field ontology, and one with a particle ontology

my understanding is that fermionic theories are easier to construct along the lines of the latter, whereas bosonic theories are easier to construct along the lines of the former

Yeah

← rejoices over recent widespread angst over sterile neutrinos, hope whole ~~standard model~~ house of cards finally collapses from them )(

and to the extent that a photon is a boson, you'd therefore describe it with a field ontology rather than a particle ontology

So the lack of a position space wavefunction may just mean that one is trying to describe the photon in an unnatural way, i.e. to describe it with a particle ontology rather than a field ontology

@vzn I wouldn't count on it

U ever see those buildings that appear to be built by termites?

termites and duct tape?

5:03 PM

but i dunno

poof :P

The SM survived neutrino oscillations, why should it care about sterile neutrinos

there is this old 2001 paper, but f*** if I know if it's right: arxiv.org/pdf/quant-ph/0102071.pdf

danielunderwood

@enumaris I was reading last night that Polchinski had to go to a psychiatrist over the proposition of a multiverse. I feel like that may be the case for more people if GR went away

if it's published in a reputable journal it has to be right

right

5:04 PM

riiiiight

no journal would ever publish anything wrong

@danielunderwood if GR went away, I'll quit physics

@vzn You are entitled to your opinion, but I think that calling the crowning experimental and theoretical achievement of partice physics a "house of cards" is a very unfair dismissal of what great successes both QFT in general and the Standard Model in particular are.

@ACuriousMind ~½ facetious. talk differently when mods are around. its like measurement altering the outcome :o

Let's partisanize physics

those in the GR camp vs those in the SM camp

vs. those in the "idfk" camp

5:08 PM

(lol) minkowskians vs antiminkowskians :P

we need to come up with derogatory names for the other camps...like and analogy for bleeding heart librals or "libruls"

just as long as you don't start throwing around words like beta

i find that crap quite annoying

beta-decay?

@vzn What? That wasn't a warning as a mod or anything, that was my very personal opinion that you are immensely overselling and overinterpreting non-mainstream approaches and immensely underselling the strength of established physics.

← has also spoken reverently of SM in here in long chat history

5:10 PM

how about we call SMers "alphabetagammas"

"The Unreasonable Effectiveness of Mainstream Physics"

3

for alpha-decay, beta-decay, and gamma-decay

@ACuriousMind dont have anything to "sell", sometimes wish there was money involved. en.wikipedia.org/wiki/Sayre%27s_law

And then we can call GRers Lambdas

for the cosmological constant

insofar as reputation/social capital is a commodity...

5:12 PM

@enumaris Call them $\mu\tau\pi$. That way, one never knows whether you're talking about physicists or a frat :P

for the muon tau and pion?

@enumaris Yup

alphabetagammatron

but that's a mix of elementary vs non-elementary particles...hmmm

@vzn If you mean to imply that a correct physical model of the world is a lesser value than money, I could not disagree more.

5:13 PM

can I stick with just one or the other

@ACuriousMind what's the use of a correct physical model of the world though? Can I eat it? Can I fill my gas tank with it? Pish posh, don't need no such thing.

@enumaris Must...resist...rant...

pokes a little harder

@ACuriousMind am not a physicist. make no money from my (sometimes passionate) interest in it. its kind of a time-sink-hole at times for me. ultimately we all want to achieve an accurate model of the universe. am playing the devils advocate/ joker/ trickster role at times. the establishment needs to be shaken out of its complacence. think a (scientific) revolution is on the horizon. very few see it coming. think it will be similar to the QM + GR revolution a century ago. on even bigger scale. etc

I'll admit, the difficulty of reconciling GR with QM is one reason I find it hard to let go of the Bohmian mindset

don't need to reconcile...just drop kick QM into oblivion

5:17 PM

On the grounds that maybe the neglect of such is what's holding it back. But I can really only wonder about that

Why not just devote time to researching relativistic Bohmian mechanics?

2

It's a reasonable enough thought.

(I even know what toy problem I'd consider first.)

sounds like a plan

Maybe

@vzn the 'establishment' invented every theory you're talking about ;)

Anonymous

5:20 PM

@enumaris There's not much use of eating or filling your gas tank either, other than keeping yourself alive and running (which is insignificant in the grand scheme of things) ;)

@bolbteppa dreaming about the New Establishment :)

I am the center of the universe, of course feeding me and keeping my gas tank filled is the most important thing

It's kind of amazing how creative 'establishment' types are and how sink or swim it is based on things like that, for people without even rudimentary knowledge of intro subjects to imply that the people who will go down in history for contribution X are 'establishment' with the connotation that brings is really something else haha

@enumaris excellent idea! except almost an oxymoron if you talk to many physicists... :|

@vzn maybe the 'new establishment' you seek is contained in here vixra.org/qgst

5:28 PM

@bolbteppa well, there's a measure of 'survivorship bias' there: The establishment types we remember are precisely the ones who succeeded

There's quite a few people who try to be establishment types who fail to make an impact and therefore are forgotten

Yeah

so I'm not sure there's a straight line from "creativity" to "establishment"

(Of course, I equally well don't think there's a straight line from "creativity" to "antiestablishment")

both establishment and antiestablishment require inspired creativity. in a way it trascends both.

How about from "creativity" to "antidisestablishment"?

The point, I guess, is that science is not so different the rest of life: it can only be understood backwards, but it must be lived forwards.

5:32 PM

Benjamin Button fan?

oh my

Nah, Kierkegaard

homepage.divms.uiowa.edu/~jorgen/kierkegaardquotesource.html

I think that quote was mentioned in the movie

ah, neat

it's a good line

Skimming those Vix articles, it's amazing how much math changes your thinking

I don't know what the "establishment" is. Is that the same physicists that so love to fight over whether string theory is worthwhile, over whether dark matter requires a modification of the Standard Model or a full extension of QFT to another theoretical framework, over the information paradox, etc.?

5:36 PM

@ACuriousMind well, I think it's a bit different nowadays versus how it was when QM was initially being worked on and codified. back then I think you could genuinely talk about an 'establishment', whereas now that's quite a bit more dicey

"establishment": there are varying degrees of actual/ perceived completion of existing theories. its ideological... aka... groupthink

coniferous_smellerULPBG-W8ZgjR

could someone review the answer by arpad szendrei at physics.stackexchange.com/questions/429510/ray-sun-heating and see my critics (both in my answer and below his answer as comments), to judge if we're both wrong or if one of us is right/wrong. thanks

@Semiclassical And yet, QM delivered. It explained what could not be explained before. Whatever resistance from the "establishment" there was, it lost.

Eh. I more had in mind Bohr et al as creating a new establishment

There's reasons why they succeeded in doing so, of course: regardless of whether one found it conceptually satisfying, QM definitely delivered on the experimental front

coniferous_smellerULPBG-W8ZgjR

"Please see here why white objects are all white" Arpad Szendrei

5:43 PM

@coniferous_smellerULPBG-W8ZgjR Arpad's answers are sometimes just rambling...

that is a particularly poor one...

coniferous_smellerULPBG-W8ZgjR

he criticize me (maybe he's right, but I think not) and then put words into the mouth of the asker that the asker never said...

I think he has a limited command of English...it seems English is not his first language

so his communication is a bit off

My point is more that the social structure of the physical sciences has changed a lot over the last few decades, and in the process it's become far more difficult to talk about 'the establishment' in an absolute sense

coniferous_smellerULPBG-W8ZgjR

just like me... but from his critics, it's clear that he understand what I wrote

the problem is not the language, it's the Physics

Anonymous

@enumaris The name sounds a bit Russian (?)

Anonymous

5:49 PM

József Szendrei

József Szendrei (born 25 April 1954 in Karcag, Hungary) is a retired Hungarian football goalkeeper. He is most famous for playing for the Hungarian national team at the 1986 FIFA World Cup in Mexico, where Hungary failed to progress from the group stage. Szendrei was a player of Újpesti Dózsa, CD Málaga and Cádiz CF. == References == József Szendrei at National-Football-Teams.com...

Anonymous

Oh, nope

I think it's a bit of both

his physics seems shaky as well

but his english makes it more difficult

Anonymous

Hungary has its own language- Hungarian....I didn't know that

o.o

You though they spoke...German?

Austrian?

Anonymous

No, I thought they used English primarily...like many other European countries

coniferous_smellerULPBG-W8ZgjR

5:52 PM

I know only one Hungarian word. kertben, which means, in the garden

hmmm...well many Hungarians do speak English

coniferous_smellerULPBG-W8ZgjR

arpad szendrei being one of them

but French people speak French and English, Spanish people speak Spanish and English...it's kinda the norm now

Of course Number Theory (Riemann Hypothesis) has the most subject-specific vixra articles on math

coniferous_smellerULPBG-W8ZgjR

i speak the 3 languages enumaris

5:53 PM

what are the chances that the Riemann hypothesis has been proved on Vixra but just nobody has seen it?

Goldbach looks more popular

@bolbteppa no shock there

This requires people to actually face things like the Gamma function so I mean I'd say they are better than the physics section

@Semiclassical my comment asking you to research relativistic Bohmian mechanics got 2 stars. I think this means a career path change is in order. :)

(Apparently not so far from skimming :\ )

5:56 PM

tell that to the job market

what job market, aren't you an academic?

eh

tbh I dunno what I want to be now

oh dear

I'd be lying if I said I didn't like research as an activity

but research as a career...not so much

hmmm

Are you not going into a post-doc?

5:57 PM

No

hmmm

So you're looking for jobs in industry atm?

I mean, maybe I'll change my mind when the time roll around next year

in principle, yeah

in practice, I'm pretty much spinning my wheels

wait, didn't you graduate already?]

yuuuuup

So you're taking a year off and just chilling?

That's cool too. :D

It took me 9 months to find a job lol, so I basically took 9 months off...

5:59 PM

The impetus for my graduating was not "you've got a career lined up, so move along" but "you've been here forever, get out already"

XD

Anonymous

@Semiclassical lol

I took the full time allotted to graduate

A guy before me had to actually petition to add 2 quarters to his allotted time I think

The big mental block I've got is my CV/resume

why's that? o.o

6:05 PM

i dunno

i've had a text draft for a while now

Anonymous

I wonder how much attention they actually give to CVs and whether it's actually worth spending hours preparing the perfect CV. People over-exaggerate on their CV all the time.

but what I need is something that's properly formatted

CV's/Resumes are actually pretty important for job search

yep

but there's a lot of filtering going on too

like just 'does this guy have this experience'? no -> auto reject

6:06 PM

I imagine that, when it comes to algorithmic filtering, the text is the important part

but when it comes to people filtering, the visual impact matters a lot

when you apply for jobs, they will ask you to fill out forms that basically re-type your resume anyways lol

lol

I went to the university's fall 'college of science engineering' career fair

Anonymous

@Semiclassical Umm...use a standard LaTeX format/template?

danielunderwood

@Blue Some people spend that much time making the perfect one customized for every single job they apply for. I can barely work through a single resume

and my main impression was "these people want undergrads or graduating engineers"

danielunderwood

6:08 PM

@enumaris I absolutely hate those sites. Especially the ones that take your resume, but use it to fill out a form and generally mess up most of it

which is not really so strange in retrospect: at a university of our size, those students are by far the majority

but it was jarring to go to multiple recruiters, whose list of majors being sought included math/physics, and realize that most of them really meant "software development"

Anonymous

I sorta feel like you should keep an eye on the R&D jobs. I presume you'd get bored pretty quickly if you get into something like software :P

danielunderwood

And I think the crazy requirements for data science positions I've been seeing is a local phenomenon. I found a posting last night that was absolutely sensible...it would just mean moving 10 hours away to Florida

Yeah, I'm not really interested in software

@danielunderwood that's the majority tho

6:10 PM

what I'm paranoid about is: "Oh, you've got a PhD? Well, we're only interested in PhD's with this niche of experience."

danielunderwood

@enumaris the majority have crazy requirements or the majority are sensible?

Or the majority involve moving to Florida? lol

Anonymous

Tbh on-campus placement is one thing that I extremely convenient. I was a bit surprised to know that that isn't the norm in the US.

majority involve moving to where the jobs are?

@danielunderwood the majority ask you to put in a resume and then fill out your resume again

Anonymous

Like you have guys from Samsung R&D visiting every year and interviews/tests are open to all

Anonymous

6:12 PM

Irrespective of degree/department

Job hunting is tough, no doubt about that lol

danielunderwood

Ahhh I thought you were replying to my other comment. Yeah I feel like job/careers/whatever sites are some of the worst sites to use

I hate em :D

danielunderwood

On the bright side, I haven't seen requests for resumes as word documents lately

PDF is where it's at

danielunderwood

6:14 PM

Yeah I always submit mine in PDF. I have seen a recruiter ask for one in word so they could edit it before sending it in. I don't know if that's supposed to be helpful or sneaky

...yeah I wouldn't do that lol

I can send a password protected word doc

danielunderwood

I'll send them the tex file of my resume :D

But I generally ignore the recruiters who don't actually work for the hiring company anyway

one thing I was surprised by: There's a Boston Scientific campus in the Twin Cities suburbs, but I didn't see any of their recruiters at the career fair

they might've been there in a different capacity, e.g. a reserved room rather than a booth

but it was still odd

danielunderwood

How hard is it to move for a new job/school?

Depends on who you are

danielunderwood

6:18 PM

I find the thought of moving to a city without knowing anyone a bit scary

In some ways, I'd be in a better position to move across the country for a job

no dependents / serious relationships, no college debt

But there's also your ability to settle into and adapt to a new place, and there I'm a lot more limited

danielunderwood

Yeah I'm definitely in a better position to do that than someone with kids or a spouse. But the closest I've done before was college and some of my best friends went to the same school

I wasn't the person that had to come home a couple weeks into college at least

Anonymous

@Semiclassical What's your feelings about consumer electronics jobs? I don't have much idea about what those actually involve

no idea either

i'm basically ignorant of that

danielunderwood

I saw some of those when I graduated. Most of them looked like sitting on an assembly line doing QA. Or at least that's how the ones willing to hire recent undergrads were

Though my school also graduated a lot of EEs, so my options for those were a bit limited

Anonymous

6:22 PM

@danielunderwood QA?

Anonymous

@danielunderwood Similar here. We have three departments consisting of EEs (around 200 students per year). Around 50% of my seniors joined stuff like Texas Instruments. I can't imagine spending my life manufacturing ICs and embedded systems tho

@Blue Quality Assurance

@Blue trying to remember what is your major?

Anonymous

@vzn Ah, thanks

Anonymous

@vzn Electronics and Communication/Computer Engineering

Anonymous

(We get to choose our specialization in the third year)

6:30 PM

@Blue seems a little strange to hear you have no affinity for "ICs/ embedded systems"... TI is one of the worlds largest mfgrs...

Anonymous

Yeah, I'm a bit weird in that respect:P. I only enjoy the extremely theoretical aspects of computer engineering and things like signal processing

@Blue lol have you heard of DSP chips? a bit revolutionary awhile back...

@bolbteppa oh, here's a preprint which you'll either love or find appalling: arxiv.org/pdf/hep-th/0702060.pdf

(I have no insight into it and therefore no opinion)

haha

(Figure 1 on page 16 is neat tho)

6:35 PM

Not sure about that non-relativistic comment at the end of 2.1

Anonymous

@vzn I did hear of them, but I don't know much. More into analog signal processing

Anonymous

I'm however, enjoying my digital logic classes very much

Anonymous

The prof is excellent and he discusses the early development of programming languages from hardware

Anonymous

(When it comes to IC designing and their internal structures, I suck tho :P)

@Blue there are probably chips that do a lot of the theory youre interested in, encourage you to look into it... if not ICs/ embedded systems then what? its a lot to "rule out" wrt your major...

6:38 PM

interestingly, the version of the guidance equation they get in the superstring theory apparently doesn't require a preferred foliation, which is the price one usually pays in order to get a Lorentz-invariant Bohmian mechanics

Anonymous

@vzn I'm actually planning to transfer to computer science/math (at least during masters)

which would be cool if true

@Semiclassical thx for the ~~ammunition~~ ref :) reddit.com/r/quantum/comments/9gbdvs/…

'Bohmian interpretation of strings'

danielunderwood

@Blue yeah that's the way that I felt. Or generally doing the same repetitive process over and over, which seemed to be more the case with larger companies

6:42 PM

@Blue =D welcome! (maybe crumbling to peer pressure on here?) lol

@bolbteppa lol. i'm sure LM would just love that

(that's not so much dripping with sarcasm as drenched in it)

Anonymous

@vzn lol. Actually, maybe. I've changed a lot in the past one year and I guess a large portion of it was due to the people on SE

someone should post it to his comments who can still post on his site :|

tbh, I wouldn't be surprised if someone has done that already: the paper came out in 2007

Anonymous

Quantum computing is like a sweet spot for me, along with algorithms in general and statistics stuff (although I don't know much stat at present)

6:44 PM

@Blue excellent decision, even though am highly biased its clear its hard to go wrong with coding in the 21st century... defn look into ML + datascience as much as possible also...

References GSW which is a plus

Anonymous

@vzn Yep :). My summer project was more or less based on quantum ML. I enjoyed it very much

Anonymous

Still working on it tho

@Blue interesting itd be cool if you wrote some of it up on the internet somewhere will read it :)

Anonymous

@vzn I'm actually writing a paper. Will share the arXiv link in a couple of months I guess

6:48 PM

@Blue great to hear... QC + ML is very hot area... a bit exotic too...

Anonymous

Extremely little literature on QML, yeah. Only a handful people work on it

Anonymous

Seth Lloyd is one of the big guys

Anonymous

There's Wittek and the Xanadu guys too

Anonymous

Rebentrost et al

Anonymous

Xanadu is doing quite well in optical QC too btw

danielunderwood

6:50 PM

@Blue Put "Quantum Machine Learning" in very big letters on your resume and apply for anything you want in the software world :D

Anonymous

lol

Anonymous

That should work

@Blue am guessing someone at google too...

Anonymous

Google seems to be a bit more focused on the experimental aspects currently ;)

danielunderwood

I have my research on mine and it usually gets a "wow that's neat", but people usually don't know enough to ask questions. Once I did get an interviewer that had a PhD in astronomy that asked me questions. That was slightly scary

Anonymous

6:54 PM

Is there any actual QML theorist in Google? I'm not sure

@Blue this i guess? ai.google/research/teams/applied-science/quantum-ai

danielunderwood

They probably have a theorist with patent rights to anything they predict or something lol

Anonymous

@danielunderwood Yeah, those sudden questions about your work can be scary. You feel like you have no idea where to start and where to stop :P

Anonymous

@Semiclassical I did see that page earlier, but I only see QML and QNNs listed as topics of research but it doesn't show who works on them. The people involved all seem to be mainly engineers

Anonymous

and software guys

Anonymous

6:56 PM

Maybe I need to dig up their publications a bit

danielunderwood

Well my work was in nuclear astrophysics and it was like "oh no he probably thinks I know a lot more about stars than I actually do"

At the one job interview I've had, I had to explain string theory to my prospective boss' boss (who is a programmer).

Apparently it worked :P

danielunderwood

I think it would have been less scary if he had a PhD in an area further away from what I did

@ACuriousMind "It worked" = "I have no idea what that guy is saying, but it sounds smart"

@danielunderwood Yeah, probably

Anonymous

@Semiclassical Ah, found one: ai.google/research/people/RyanBabbush. Not exactly QML however. General QC algorithms for physical implementation

danielunderwood

7:00 PM

Or he was secretly a string theorist making sure you were legit

0

Q: Two similar-looking mechanics questions are not duplicates

This question about which way a bicycle will move when the pedal is pulled horizontally was closed as a duplicate of this question about how a yoyo will move when it is pulled horizontally. They are not the same question, because a yoyo is a rigid body, while a bicycle is not even close. The cor...

Anonymous

I don't actually understand why they call it the Quantum AI lab

Anonymous

I've not seen them publish any papers on artificial intelligence (AI)

@danielunderwood Nah, he's got a diploma in computer science. He later said he just wants to hear applicants talk about something they claim to know well

Anonymous

@DanielSank Why is your group/team named as Quantum "AI"? :)

danielunderwood

7:04 PM

Ahh I could see that being a good thing to do in an interview

a question I should know the answer to: Is it obvious why $e^{-2\pi i J_i/\hbar}=(-1)^{2j}$ for a spin-$j$ particle, regadless of which component $J_i$ you look at?

@Semiclassical $2\pi$ is a full rotation.

Or "half a rotation" in the case of a fermion.

Hmm. Yeah, that'll do

Either way, doing a full or half rotation is the same regardless of which way you rotate

The overly-complicated answer I had was that all such operators $e^{-i \vec{\theta}\cdot \vec{J}}$ with $|\vec{\theta}|=2\pi$ were unitarily equivalent, so it should suffice to consider the case of $e^{-2\pi i J_z/\hbar}$

In which case it's diagonal and easy

But that felt like I was forgetting something simple

(it doesn't help that the notes I'm reviewing are full of typos)

Anonymous

7:50 PM

~~Wtf~~ What is up with people writing their bio in the third person: quantumcomputing.stackexchange.com/users/356/brian-r-la-cour. I don't know if it's a cultural thing, but looks extremely weird and creepy to me

Anonymous

Hmm, so third person professional bio is a thing: nectafy.com/how-to-write-a-professional-bio

webpage blocked

lawl

8:22 PM

@Blue The idea was that quantum annealing would help problems in the "machine intelligence" class.

@JohnRennie How come you get to make jokes with sexual innuendo, and even have it pinned to the star board, but when other chat room users make a political or sex joke they get chastised?

@Blue A lot of academic bio pages are written that way.

Anonymous

@DanielSank Ah, I see. I still think a more descriptive name could have been chosen (especially because annealing is probably not the sole topic of research there) :P Although "AI" inserted anywhere makes it cooler for the public!

because double standard o_O :(

Anonymous

@DanielSank I've seen some of those before, yeah. Nevertheless, makes me a tad uncomfortable XD (and more so when I am almost certain that it's not someone else who has written the bio for that person)

dang, my package is now expected to arrive tomorrow -.-

delayed by 3 days cus they kept bouncing it from facility to facility :(

Anonymous

@enumaris Reminds me...I finally ordered a copy of Kernighan and Ritchie

8:35 PM

which book is that?

Anonymous

Should be a nice experience to read a so-called classic

Anonymous

@enumaris It's not a physics book :P

Anonymous

A popular C book

oh

Anonymous

By two of the original founders

8:37 PM

what's a C book?

like for programming C?

Anonymous

Yeah

Anonymous

I'm trying to improve my hold on the nitty gritty details. The Schaum series got boring after a while

Anonymous

Very few books actually go into the details

I see

so you want to program in C?

and not C++?

is there a reason?

Anonymous

C++ uses a lot of abstractions. I already have Python for that

Anonymous

8:42 PM

Not that I don't want to learn C++ well. But C is like the building blocks of modern languages

Anonymous

C++ also has STL which is nice

Anonymous

Oh, and I don't know much about the Python libraries. There are so many! Should probably read Summerfield

C++ and Python are quite different...

Anonymous

C++ is faster :P

well, you'd be surprised how optimized python libraries are...and when they need speed they just use a C++ backend anyways

danielunderwood

8:49 PM

Python is a bit more flexible. Although I think C++17 is bringing in support for metaprogramming, which is interesting

It's the reason why deep learning libraries can exist in Python...if you had to use pure python it would be way too slow

danielunderwood

Or brought in I suppose I should say. Though a lot of things are barely getting to supporting C++11 these days

Anonymous

The thing is, I'm not really that much concerned about speed of execution nor ease of development at the moment. What I'm interested is optimization of memory usage...which most languages seem to hide under bulk of abstraction

danielunderwood

You can control your register usage if you go all the way to assembly!

Maybe there's a way to do that elsewhere as well...I've never had the need to do that

Anonymous

Yeah, planning to pick up assembly :P

Anonymous

8:53 PM

I was sorta motivated by an SO user who managed to speed up one of my program 1000x compared to the original naive version I wrote

danielunderwood

Also I agree on 3rd person bio being odd

Anonymous

Things like, calling rand() utilizes a huge part of run time

Anonymous

It can be cut down many fold

danielunderwood

I'm amazed that people can come up with stuff like this en.wikipedia.org/wiki/Fast_inverse_square_root

How would you improve that other than queuing random numbers?

Anonymous

@danielunderwood Some bit shift tricks are faster. Helps when you have to call random numbers 10^3 or more in a program

Anonymous

8:59 PM

It's pseudo random tho....but good enough most of the time

danielunderwood

Ahh I've actually avoided random numbers a fair bit somehow

And searches and sorts

Anonymous

My project was based on random matrices. So I had to make a lot of random number calls :D By lots I mean like $10^9$ times in a single run

danielunderwood

Ahh. I actually did some detector simulation work at one point that I assume needed a bunch of random numbers. I just didn't happen to be the one generating the random numbers

ML Algorithms are very heavily dependent on random numbers

for initialization

(some)

danielunderwood

I guess I'm very dependent on random numbers. I just don't personally have to use them

Aside from the handful of times I've used pure tf and had to initialize the values

9:05 PM

I'm sure you've used random.shuffle()

that's based on random numbers obviously

danielunderwood

I'll revise my statement. Evidently I use random numbers every day and just don't think of them

:D

danielunderwood

But if random numbers are expensive, I could just not shuffle my data! :D

Anonymous

On the other hand, cryptographically secure randomness is a whole field in CS

Anonymous

Bit shifts are not secure

9:08 PM

just use quantum randomness

it has the word quantum in it

so it's guaranteed to be better

danielunderwood

This kind of seems like randomness in my work

Some engineer out there has solved P=NP and it's locked up in an electric eggbeater calibration routine. For every 0x5f375a86 we learn about, there are thousands we never see.

Anonymous

@enumaris Heh, those can be time consuming :P

quantum

guaranteed to be better

danielunderwood

why stop there when you can have a quantum machine learning algorithm for randomness?

sounds good

Quantum A.I.

trademark enumaris

I'm gonna be rich

danielunderwood

9:11 PM

I'll call it QuaMLAR

Anonymous

Relativistic String Quantum Blue A.I. - that could be good name to sell myself

danielunderwood

Nooo someone already owns quantum.ai and nothing is there

I was going to buy it and put something stupid there if not

lol

url squatters man

danielunderwood

Forreal

I have an idea and I go to look up the domain and it's always just someone trying to sell it

At least I wasn't offered the opportunity to buy quantum.ai for some outrageous price

just wait a couple years

danielunderwood

9:17 PM

Ahh right. Why sell for an outrageous price today when they can sell for an absurd price in a couple years?

indeed

danielunderwood

The solution is to just invent the internet. Then you don't have to worry about url squatters home.cern

Anonymous

And then there were people who bought google.com for $12 and even returned it...(don't remember if that was for real)

Thank you for the clarification, Emilio. I will adjust the post. — stagerf 48 secs ago

well, that's a refreshing change

Anonymous

9:35 PM

Judging from the post formatting and language, he/she doesn't look like a new user to me. I'm guessing it's a previous SE user with an alternate account (or one who has returned after a hiatus)

10:19 PM

hmmm

« first day (2877 days earlier) ← previous day next day → last day (2054 days later) »

Transcript for

Sep '1820

The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile