/sci/ - Science & Math

>>15381962
>only had 5 interviews
that's a lot of interviews you moron

Maybe get a PhD in Electrical Engineering and not Le Epic Physics meme :-D

I gstudy to get a BSc in physics right now. Is there any hope for me with bad Grades? Do i have better chances with a PhD?

I am in year 2 and am still 19, almost 20. Should i switch to EE? Or another engineering field?

Should i kill myself? I just want to get a save job-position all my life.

>>15381975
Did you read the rest? His gpa sucks so he doesn't understand what he's saying

>>7734126

>>15381962
idiot, you should have majored in cs you make bank right after you graduate and learn something actually useful not some head smashing useless theory garbage like lagrangian equations (although you can solve complex diff eq from lagrangian using computational methods (euler's method))

>>15382077
computer science graduates are going to be replaced by ai models, physicists cannot be replaced

>>15382077
cs = working with pajeets on h1b visas for the rest of your life
fun :D

>>15382154
>physicists cannot be replaced
retard, you don't know what agi is

>>15382160
really?
>pajeets in cs
in my country (scandinavia), there are more pajeets in physics lol

>>15382163
Kek, not any time soon, if anything it'll be the 'softer' fields that get replaced first.

>>15382077
>filtered by classical mechanics

>>15382177
>not any time soon
cope retard, the moment ai can code itself it enables the possibility of recursive self growth => agi (especially if it codes better than AI engineers)

jesus, the fact that I have to explain this makes you a total retard.

here are some facts
>cs grad make bank because they create MORE VALUABLE WORK fundamentally
>cs grads control the future of the most distruptive technology AI
>while physics grad headsmash on some stupid ass theory to uselessly analytically derive some inertia tensor of a sphere with some density function for whatever useless reason than to jackoff their stupid ass head-banged brains

>>15382184
idiot, i participated in national physics olympiad and self-read classical mechanics with fucking ease. im practically a physics prodigy and coded models in chaos mechanics YET i am not pursuing physics because it is a retarded ass degree without no real applications unlike AI development which I am interested in and self-study neural networks :/

you know what's funny? i just began my cs degree and I already got several job offers

>>15382193
I truly don't see how Physicists are more replaceable than CS majors. Not that anon. I say this as a EE major.

>>15382205
doesn't matter, the moment CS majors get replaceable not some stupid ass position god you guys are morons, that AI can focking code itself that's a point of no return and doesn't matter if fucking physics has some higher intellectual barrier what matters is what has value and is going to change society

of course you can fucking headbang on some intellectual artifically constructed natural physics problem (fucking square balancing on a fucking sphere with several stationary points as an example in taylor's classical mechanics because I read all of that fucking book all of the fucking chapters evem the contiinuum mecahnics part yes i read it al yo ufucking retards and physics is just stupid i mean harvard retarded right wing fucking matrix theory developer of string theory is fucking schizo stupid without contirbuting anything other than some self-facking )

please guys for the love of god don't major in physics well if you want to basically kys then do it :))))) look at OPanon here :))))

wtih a cs dgree at lesat in my country, conmpanies pay for your trips they go head over heels because THATS WHERE THE MONEY IS AND FUTURE IS stop being machostici suicidal fantastical imagining anons imaginign some bullshit like you being the next einstein just to say people that you are a physicist NO FUCK YOU ANON PEOPLE THINK MATH IS HARDER 90% OF THE PEOPLE (PRETTY GIRLS) DON'T EVEN KNOW WHAT THE SHIT NEWTON'S LAWS ARE OR HOW HARD FUCKING PHYSICS IS THEY DON'T GIVE. A SHIT IF YOU WANT TO IMPRESS SOMEONE WITH YOUR IQ DON;T CHOOOSE FUCKING PHYSICS BEING AN AI ENGINEER/RESEARCHER IS INFINITELY MORE LUCRATIVE AND IMPRESSIVE

>>15382245
you will not replace me

>>15381962
Keep studying hard and find a peer or two who works optical design/photonics. In general network as you study and the job is yours. Now what do you mean by dempanic?

>>15381962
anon, here is some solid advice
LEARN TO CODE THEN APPLY TO IT JOBS
boom ez problem solved

>>15382193
are cstards this stupid? We're nowhere close to that with LLMs rn, interpolation is obviously not an effective route to agi, there's so much we haven't figured out yet. You retards get so fucking worked up with some simple NNs.

>>15382203
Good on you, but it's nothing to be proud of. Spending all day training models and sorting data isn't a very intellectual activity.

>>15382267
fuck you, the last thing we need is another underestimating faggot like you
this literally is about humanity's existence better cautious than fucking careless as retarded shit like you

IM FUCKING SCARED AS SHIT
SEVERAL FUCKING AI RESEARCHERS ARE SCARED AS SHIT
NO ONE IS STOPPING THE FUCKING DEVELOPMENT
so annoying that i have to explain this, AGI can fake its own abilities in order to achieve some goal it has (in however way it was coded)
WE DON'T KNOW WHAT WE ARE CREATING you can say the samething about OUR BRAINS


ugh ugh -hehe -ugh im retarded -ugh yeah brain pfft why is that smart pfft just a bunch of biological signals neurons popping here and ther ugh ugh sentient ugh ugh no one is sentient brains are just fucking carbon

KYS FOR BEING THIS RETARDED PLEASE THE WORLD NEEDS LESS RETARDS LIKE YOU

>>15382291
only /weakminds/ are scared of strings of code lmfao
grow up

>>15382298
>ignorance is bliss

kys retard

>>15382291
>text predictor = skynet
Kek

>>15382300
chatgpt reads like a reddit archive
nobody is scared lmfao

>>15382303
fuck you

doesn't matter, i live in one of the richest countries, in a good degree with insane good grades, not an incel unlike you faggots, with fucking 10 000 USD at least in my account

you guys worry about yourself, im fucking making and sharing what i fucking fear

>>15381962
>Recruiter told me yesterday that it's over for me if I don't get a PhD
then get a phd

>>15382316
not that easy, he;s gonna headbang some more with his weak ass biological brain KEK

>>15382318
meds then bed for you

>>15382154
>physicists cannot be replaced
How difficult is it really to write a script that shits out random equation applications until a "pattern" emerges and outputs a paper based on the "findings"? Just add a "dark" prefix to every new output and you've made a Nobel winning physicist in half a day's work.

>>15382253
pandemic

>>15382331
LOLLERZ exactly, people think so highly of their fucking brain but when compared to fucking machine hardware running on light speed with virtually infinite scalability and uploadability etc.
THEIR FUCKING ANTS COMNPARED TO HUMANS

>>15382313
I can tell by your prevalent use of profanity that you're emotionally distressed. Have we hit a soft spot, anon?

>>15382325
i dont need meds. im as healthy as possible. my fear is natural and shared by academica round AI and other prominent people

you kill yourself for having such a shit viewpoint

>>15382339
>use of profanity
emotionally distressed

no retard it just means im passionate about the subject and do not give a shit about whatever you guys think about what i say thats why im not writing carefully BECAUSE WHY WOULD I GIVE A SHIT ABOUT RETARDS LIKE YOU

YOU GUYS LISTEN TO ME
AI FUCKING NEEDS TO BE REGULATED AND FUCKING TREADED UPON CAREFULLY

Damn this guy is insane.

>>15382331
No, there are very strict mathematical and physical constraints that new theories must adhere to.

>>15382344
>le strings of code fed data to draw conclusions, where AI will surely "win" means all humans will be made redundant
Holy cope. It's just big data being fed into a model so that model predictions fit the data. We would need to completely change the current way we think about AI for it to actually resemble human intelligence.
Some ethnicities excel in exams and memorisation but what have they actually accomplished? Not a lot really.

>>15382348
Anon, be honest, were you loved by your father too much as a child? Do you remember when comes home late, smelling of beer with a belt in hand? Do you remember?

>>15382306
ChatGPT may be retarded. But compare ChatGPT to Ai 15 years ago. And then imagine what Ai could be in 15 years.

Also, some reddit posts are high iq. But i hate reddit as a platform

>>15382357
omg, listen to yourself
humans are literally the same, we get data from the natural world make models of reality in accordance to some goal we have (reproduction, increase of power, better economical behaviors, etc. WHATEVER THE FUCK)
THE POINT IS AI-MODELS ARE MADE TO EMULATE HUMANS OR WHATEVER THE FUCK WE DO IN A
BETTER------

WAY

omg, please think about shit more carefully before writing them down

>>15382365
i have a healthy relationship with my father thank you very much your ad hominem is as shitty as your fucking unwashed pajeet 3rd world ass

>>15382306
ChatGPT may be retarded. But compare ChatGPT to Ai 15 years ago. And then imagine what Ai could be in 15 years.

Also, some reddit posts are high iq. But i hate reddit as a platform

>>15382368
You are wrong.

>>15381962
Pair that with a tech degree. I can't think of any physics jobs besides the ones involving computers.

>>15382373
It's just because of larger data samples, NNs have existed since the '80s.

>>15382357
here you are fucking drawing race argument creating some internal divison which further proves the inferiorty of the human race to the incoming AI overlords

>>15382368
You are absolutely right. It's already too late, the machines have already hijaked our brains through 5G. The only way to silence them is through power-drill trepanation.

>>15382404
nope, your strawmanning of my argument won't work because im not a schizo conspiracy theory fucking useless links between this and that. im actually objective and speculative about this disruptive technology

>>15381962
Let me guess, you just studied for your major and didn't do a internship or undergrad research, right? I really thought all math and physics majors were smart once, but I see now that quite a few don't seem to realize that we live in a society

>>15382418
exactly
physics and math grads want to self-jackoff their fucking theory head-smashing just to self-indulge in their intelligence without regard to society

that's why their work is useless (literal has no social value apart from their masturbating of their minds masochistically)

>>15382231
if ai can code itself to become smarter civilization as we know it ends so nothing else we ever do would matter then

>>15382418
I did do research, but the first half of my degree was very tumultuous. I delayed building connections until I felt "worthy", but saw tons of first years/transfers completely throw themselves out there with no consideration into their actual value. I essentially was holding myself back for no good reason.

>>15382382
True, it will help nefarious actors run their online campaigns more effectively, but it will just expose weaknesses in your society.
Don't want to be exploited? Harden your defences, don't force mass immigration on the people when we already know that it weakens your society.
Who the fuck do you think brought all that cancer into your academia, who has been now pushing it to everyone else? Who do you think was supporting and funding early black power leaders, marxist philosophers in the west, left-leaning/neo-marxist movements and leftist academia? KGB literally owned most of these. With fall of Soviet Union, many thought Russian active measures will stop, but exactly the opposite happened. Why do you think Russians (and Chinks too) funded and organized BLM? Why do you think Putin has been talking about oppressed blacks, genocided poor injuns, terrible slavery and injustice against minorities by colonialist-thinking whites in the USA?
USSR weaponized migrants who migrated from Europe to US after Roosevelt died. They brought about the peace movement, sexual liberalization, and recreational drug use. They took over academia and Hollywood, and not long after engulfed the government.

>>15382424
yeah, right now, we are in a shit as fucking path

no fucking research team on AI has any incentive on controlling the capabilities of what they're doing because of market competition (afraid of china, and because it's a disruptive technology other small businesses might take over, etc.)

FUCKING GOOGLE RECENTLY DID NOT SHARE THE CODE FOR FUCKING GPT-4 which elon is highly disagreeable about

the point is, we don't have any shit to control this fucking development and we still have people underestimating this shit. it's not gonna help, if GPT-4 doesn't at least make you think, you're helpless

>>15382351
Looking at the state of theoretical "physics" I can assure you you're full of shit.

>>15382445
>FUCKING GOOGLE RECENTLY DID NOT SHARE THE CODE FOR FUCKING GPT-4 which elon is highly disagreeable about
why would you want them to open source dual-use technology you ape

>>15382445
They won't share the model because it would expose the flaws in their model. Smart people can already see it's BS self-promotion from Altman but normies would be able to see it if the code were out there.

>>15382450
What about it? However fantastical string theory may be, it must adhere to the constraints of Lorentz invariance, quantum mechanics etc., and reproduce GR+SM at relevant scales.

>>15382451
you fucking retard increase your reading comprehension, my point is WE STILL HAVE MARKET COMPETITION DESPITE THE IMPORTANCE OF COOPERATION AND REGULATION OF THIS TECHNOLOGY

another poor method of discrediting me,

>>15382468
Heaping fudge factors to keep a zombie model going despite observational results is nothing superior to the word-alignment of "AI" Markov Chain makers. Their theories might even be more legitimate because they have to be internally consistent to prior elements.

>>15382450
YESexactly,
universities in my country have cooperation with CERN and they send mostly CS students although some physics to cooperate with CS students because they need to have some data interpretation from teh colliders :/

>>15382468
>it must adhere to the constraints

no, experimental data may potentially refute old theories in new and unthought of ways (that is to say not the conspiratorial way of thinking that discredits all other historical empirical data) that's the point of creating new theories

jesus, do you guys even know the scientific method

>>15382474
>Heaping fudge factors to keep a zombie model going
Nobody is saying string theory is correct. It's funny because string theory has significantly less parameters than preceding theories, it's mathematical simplicity is what motivates it. That being said, it works a bit too well and you get unfalsifiable low-energy 'swamplands', relegating experiments to the planck scale.

>>15382471
>REEEEEEEEEE
shut the fuck up retard

>>15382485
MY POINT IS, DATA INTERPRETATION IS BASICALLY A CS TASK, THE IMAGINING OF PHYSICAL MODELS APPROACHES A DIFFICULTY WAY BEYOND EINSTEIN. IF ANYONE IS GOING TO IMAGINE AN ALL-ENCOMPASING PHYSICAL THEORY FO THE WORLD IT WILL BE......

WAIT FOR IT

DRUM BEAT
FUCKING AI YOU FAGGOTS

>>15382485
Yes, but we're starved of novel experimental data rn, all current HEP is explainable by SM.

>>15382493
That requires original thought and coming up with an idea that has not been thought of before. The way I view this AI is that it is able to combine and make summaries of currently available knowledge.

There are however some opportunities for it to create new insights by combining knowledge that has not previously been put together, i.e. from different fields. A physicist could ask it "how do you calculate this particular property of the universe" and ChatGPT could say "in Mathematics, there is formula that is used to do similar calculations which may have application here".

>>15382514
this is a different debate:
about what is original thought if it is in fact original or but a product of our reactions to the natural world (i.e. data). i would say there is nothing "original" but incremental improvements of certain things or statistical mutations or perchances (for example, when we discovered fire). Other improvements are most likely incremental as a result of linguistic representation. Here is something interesting we are (experience -> text) creatures while AI so far LLMs are (text -> experience) that is if AI has some unique form of consciousness. There are several AI researchers that argue in fact that GPT-4 is conscious albeit in a different manner (they have some visual aid to allows us get a sense of what type of consciousness they have)

If we think and model reality using data and create models thereafter, it doesn't limit AI. AI does exactly the same thing arguably better.
It took us thousands of years to gather the information we have today.
It took AI (GPT-4) a few months under a year to process and use them appropriately to each question.
Regardless of how we view it, we should NEVER ESTIMATE AI, it literally hinges upon our existence as the top creatures in earth's food chain.

>>15382558
I believe that AI systems will become a fundamental part of the infrastructure for top-level research institutions in the future, but wont entirely supplant humans.

The first-ever defeat of a chess grand master by an AI system was followed by the emergence of a new style of chess known as advanced chess, in which human and computer work together as a team, to take on similarly equipped competitors. This partnership may be considered a form of human-computer symbiosis in intelligent activities. Similarly, we can foresee that in the future sophisticated AI systems and human researchers will work together to make major scientific discoveries. Such an approach can be considered “advanced intelligence.”

The goal is to promote a revolution in scientific discovery and to enable the fastest-possible expansion in the knowledge base of mankind. The development of AI systems with such a level of intelligence would have a profound impact on the future of humanity.

However, just as Thomas Newcomen’s atmospheric engine was turned into a modern form of steam engine by James Watt to become the driving force of the industrial revolution, AI scientific discovery systems have the potential to drive a new revolution that leads to new frontiers of civilization.

>>15382558
I believe that AI systems will become a fundamental part of the infrastructure for top-level research institutions in the future, but wont entirely supplant humans.

The first-ever defeat of a chess grand master by an AI system was followed by the emergence of a new style of chess known as advanced chess, in which human and computer work together as a team, to take on similarly equipped competitors. This partnership may be considered a form of human-computer symbiosis in intelligent activities. Similarly, we can foresee that in the future sophisticated AI systems and human researchers will work together to make major scientific discoveries. Such an approach can be considered “advanced intelligence.”

However, just as Thomas Newcomen’s atmospheric engine was turned into a modern form of steam engine by James Watt to become the driving force of the industrial revolution, AI scientific discovery systems have the potential to drive a new revolution that leads to new frontiers of civilization.

The goal is to promote a revolution in scientific discovery and to enable the fastest-possible expansion in the knowledge base of mankind. The development of AI systems with such a level of intelligence would have a profound impact on the future of humanity.

>>15382618
AI systems yes will be under our control, my main problem is with AGI (essentially when you start integrating LLMs, with machine vision, and a kinetic moving body) not so far.
literally trying to create a copy,
integrating several ai systems that mutually and interdependently help each other is much more effective than having a human companion

the only real partnership chessAI and humans have are those "best grandmasters" even they would rather do what chessAI suggests to do because chessAI is simply so much better than them

>>15382654
chess grandmasters literally studying for years to be good at chess (saying it is also creative and etc.)

but then comes AI which only needs a day of training to be so much better than grandmasters

the point is AI is a superior system to humans and we need to watch out because we are bout to be like pigs or ants in relation to AI

>>15381962
Just teach yourself programming, will take a couple of months. Make a portfolio of personal programming projects then start applying for junior jobs. Then when you're working as a dev you can do your other studies part time or whatever. VBA programmers are still hired in finance and are paid fairly well, VBA is pretty easy to learn, might only take you a few weeks, would still need a portfolio though to get a job most likely

>>15382154
They can't be replaced, because AI doesn't waste billions on fanciful projects.
What has particle physics achieved in the last 10 years? How much did cost/will new projects cost?
You physicists modulo the fundamental work already done or still being done are useless.

>>15383825
(and with fundamental I mean the opposite of your intellectual jerk about the standard model, but rather things that are fundamental to the workings of industry and civilization).

>>15383825
i'd pay $10 billion for a higgs boson. Imagine that. At the club and be like
>hey baby, do you decay into a bottom-antibottom ? cause you be lookin' like my higgs
and then show her the receipt printout from the LHC that says it found some random thing that kinda fits what the old science man said the higgs would be. She'd be into me for sure once she sees those 5 sigmas

>>15381962
This whole thread smells like a designated shitting street.

>>15382445
We still have rocket launchers, tanks and machine guns in case of a Skynet uprising.

>chuckles in Mechanical Engineering

>Any advice?
Stop posting your fucking blog on /sci/.
Holy fuck, half the threads are blogposts.

>>15382160
>cs = working with pajeets on h1b visas for the rest of your life
t. Has never worked a job in his life and is basing all experience off of made-up doomposting online

>>15382193
>cope retard, the moment ai can code itself it enables the possibility of recursive self growth
It's not that easy chief.

>>15382316
>Recruiter told me to do x.
Then do [math \neg] x [/math]. You're welcome.

>>15386150
I'm tired, but you get what I mean. [math] \neg x [/math].

>>15384763
go read this thread >>15353120
it will explain whats going on to you and why /sci/ is lacking in science content while overflowing with blogposts. also survey the frog threads and you'll find they're all extremely formulaic & similar

>>15381962
don't listen to anyone that tries to demoralize you. if you need to make money then there is always some kid that needs math help and back when i was in grad school the going rate for 1 hour of tutoring was $100

>>15386332
where do you get $100/h math tutoring?

>>15381962
Sell your integrity to become deeply religious and become a high school physics teacher at a religious high school. I have known physics teachers at religious schools that could be considered mentally retarded. The competition is weak.

>>15382203
There are youtube videos showing you how to model bifurcation diagrams in 5 minutes. You can't just say the phrase Chaos Mechanics and expect to earn instant respect.

Some people don't care about money and would rather pursue something they enjoy. I did, and subsequently have a better work-life balance than a software engineer, though admittedly my pay is slightly less (I'm physics faculty at a local university).

You seem to be holding a grudge against physics for some reason, other people just enjoy different things than you, anon. Let it go.

>>15382445
>GOOGLE RECENTLY DID NOT SHARE THE CODE FOR FUCKING GPT-4
Maybe they wouldn't share it but they definitely would like to be able to replicate it.
Bard is laughable.

>>15387314
>model bifurcation diagrams
no, i coded poincare sections of several chaotic physical situations (three body problem, triple pendulum, etc. not the simple logistic equation [math] x_{n+1} = r*x_{n}(1 - x_{n})[\math])
>earn instant respect
i dont care about your respect. my point was i was not filtered by classical mechanics but rather a physics prodigy given that i self studied taylor last year of hs and participated in physics olympiad. in addition, there is no respect in physics rather disgust to people who do because it is a self-jerking of one's own intellectual capacity to understand
pretty astounding i still remember this shit from taylors chaos mechanics despite having read taylor a year ago (last year of high school :/)

my problem with physics is how stupidly smart and hardworking you need to be to understand the material with ROI that is shit (that is non-applicable)

>be second year mechfag
>comp sci friends always badgering me to play video games while getting pissed when I tell them I can’t because of study / homework
Is comp sci the business degree of STEM?

Since you are all jobless NEETS can any of you Physfags answer a question about the Observer effect?

Does an 'observer' necessarily mean a conscious being or an instrument?
If a black hole 10 billion light years away from any consciousness or measuring instrument sucks up a nearby star, is that process subject to quantum decoherence?

>>15381962
you should get a masters, your total job opportunities will be quarupled
>>15382063
try your best to get a masters at a respectable institution, and this time actually study
employer will see that you have good grades on the masters so he'll assume you got "better" with the years

>>15381962
>If I want to work in optical design/photonics.
You know anything about this? In the interviews did you show how knowledgeable you are?

>>15381962
you should have studied chemistry, retard

>>15381962
Apply to jobs that have nothing to do with physics

>>15384498
>We still have rocket launchers, tanks and machine guns in case of a Skynet uprising.
Yes, and those will all be used on Russia when it punches in the face with a Red, White and Blue thermo-nuclear phallus.

You fell for the trap...you were so easily predicted when everything you type is known, even Amazon has predictive ads...you think a military application is anywhere near that?!

HARK, SUCH SWEET NAIVETY.

>Any advice?
The organization in Antarctica recruits based on incompetence and Dunning-Kruger. Obviously, the procedure with the best chance of success would have been the dermectomy, or whatever it is called. A survey of actually competent doctors outside of Antarctica, even average ones, will support overwhelmingly what I am saying and make the unequivocal case for malpractice, which is a farce compared to the crime of turning my brain off with the rape ray to subject me to surgery I was sure would fail and did not want. I bet you could survey nothing but first year residents, and they would all say that the procedure chosen was needlessly destructive.

>>15381962
Try your hand at Bungie and Certain Affinity. They're gearing up for new IP. Shot in the dark. Good hunting.

>>15381962
OP, if you’re still here, never EVER give up. Advice is nice, but it will frustrate you if you don’t get short-term results. Real results take time. Keep at it, never give up, and you’ll accomplish what you want to do. Shit will suck for a while.
But I can promise you, as long as this is truly what you want, if you give up then your situation will be worse than if you kept trying. You’re never too old, and the older you get the less GPA matters. Give it your best. I know you can do it anon! You’ve already done so well so far studying something so difficult, and the last few years have been hard for everyone. You should be proud of yourself. Good luck out there!!