[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math


View post   

File: 430 KB, 1600x2244, __patchouli_knowledge_touhou_drawn_by_nikorashi_ka__6ef8819a79c1fa26a2b0fe4959d9c183.jpg [View same] [iqdb] [saucenao] [google]
12592403 No.12592403 [Reply] [Original]

Patchouli Patchouli Patchouli edition.

Formerly >>12570353

>what is /sqt/ for?
Questions regarding math and science, plus related advice requests.
>where do I go for other questions and (advice) requests?
>>>/wsr/ >>>/g/sqt >>>/diy/sqt >>>/adv/ etc.
>how do I post math symbols (Latex)?
https://imgur.com/MDiglsS.png
>a plain google search didn't return anything, is there anything else I should try before asking the question here?
https://scholar.google.com/
>where can I look up if the question has already been asked here?
>>/sci/
https://boards.fireden.net/sci/
>how do I optimize an image losslessly?
https://trimage.org/
https://pnggauntlet.com/

>where can I get:
>books?
https://spoon.wiki/Books
https://stitz-zeager.com/
>articles?
sci-hub.st
>book recs?
https://sites.google.com/site/scienceandmathguide/
https://4chan-science.fandom.com/wiki//sci/_Wiki
http://math.ucr.edu/home/baez/physics/Administrivia/booklist.html
>carreer advice?
https://sciencecareergeneral.neocities.org/
>help with calculus?
https://spoon.wiki/WolframAlpha
>charts?
https://imgur.com/a/JY6NNeL
https://imgur.com/a/0qDEgYt
>tables, properties and material selection?
https://www.engineeringtoolbox.com/
http://www.matweb.com/

General advice for asking questions here:
>attach an image (animal images are best. Grab them from >>>/an/)
>avoid replying to yourself
>ask anonymously
>remember to check the Latex with the Tex button on the posting box
>if someone replies to your question with a shitpost, ignore it
>avoid arguing with Yukarifag
>do not tell us you came from whatever the fuck board, /pol/ in particular
>do not mention how [other place] didn't answer your question so you're reposting it here
>If you use j for the complex unit, put a ¿ somewhere in your post or use emoticons I will automatically ignore your question. I don't actually know about everyone else, but you shouldn't assume they're too far off about whatever random things they dislike

>> No.12592442
File: 419 KB, 812x1143, __patchouli_knowledge_touhou_drawn_by_ka_marukogedago__ce2fe306e064317d4e77afc870cd8416.jpg [View same] [iqdb] [saucenao] [google]
12592442

Unanswered questions:

Math questions:
>>12579604 [Might have been answered by the questioner in >>12580029 , I dunno how to read this.]
>>12590735
>>12591723
>>12591871

Physics questions:
>>12579059
>>12579494 [Very neat question.]
>>12584691 [If we're talking maths, the answer given is correct. I recall there being a convention in QFT where you moved the path to the left or something tho, which might be what anon actually wants.]

Chemistry questions:
>>12580872
>>12584356

Biology questions:
>>12588008
>>12590695

Medicine questions:
>>12590348

/g/ questions:
>>12570984

Stupid questions:
>>12570556
>>12570934
>>12579287
>>12581391
>>12586793
>>12586841
>>12587150
>>12588003
>>12588356
>>12591436 [I'm still doing my bachelor's.]

>> No.12592456
File: 306 KB, 2048x1370, __patchouli_knowledge_touhou_drawn_by_samayoi__d452058022bf20d73b789b68d6f554db.jpg [View same] [iqdb] [saucenao] [google]
12592456

>>12591871
Consider [math]f(z) = z + 100[/math].
Then [math]\sup _{|z| < 1} |f(z)| = 101[/math] but [math]\sup _{|z|, |w| < 1} |f(z) - f(w)| = 2[/math].
What condition did you forget to add to your post?

>> No.12592493

Because electron orbitals have irregular shapes, do all atoms, even single hydrogen atoms, lack spherical symmetry? Do gravitational tidal forces have a tiny but nonzero effect on them?

>> No.12592696
File: 56 KB, 817x139, qtddtot.png [View same] [iqdb] [saucenao] [google]
12592696

How do

>> No.12592704

>>12586793
extract urea, gunpowder (nitrate), white phosphorus
taste it to see if you have diabetes
Maybe scatter it around your veggie garden to scare off prey animals

>> No.12592714

>>12592493
>electron orbitals have irregular shapes
orbitals, like p- and d- orbitals, are exaggerated and shown in isolation for illustrative purposes. In reality, the sum of all 3s, 3p and 3d orbitals makes a perfect(ish) sphere (although perturbed/deformed by their environment)

Orbitals are just allegories for how angular momentum behaves in practice, for the purpose of mathlets who need an intuitive tool for complex chemical reactions

>> No.12592717

>>12592714
When perturbed into not a perfect sphere, do gravitational tidal forces apply?

>> No.12592740

>>12592717
gravity affects everything, dumbass

>> No.12592787

>>12592740
Yes, but is there a tidal component or are all parts of an atom affected equally?

>> No.12592979
File: 926 KB, 1797x1024, emotions.png [View same] [iqdb] [saucenao] [google]
12592979

Retard here, or at least guy stuck with a problem that makes me feel like one.

Doing ML Regression; have a huge vector of like 3500 variables, which need to predict a single output value.

However, the error I'm seeing is absolutely terrible. Zero predictive ability from any of the input values, despite trying tons of different network topologies and setups.

Dataset is enormous.

Do I keep finding/trying to add more variables? What else can I try?

Is there some kind of meta-guide I can follow on how to troubleshoot this scenario?

Client is counting on me solving this and I would love to keep my job.

>> No.12592992

>>12592979
Average out data monthly/weekly and then try again.

>> No.12593014

>>12592992
The input vector already includes a ton of different summary metrics of the daily data.

How would weekly/daily help?

>> No.12593026 [DELETED] 

>>12593014
Your data set may have funny short term variations which hide the long term correlation.
Consider for example the series
0 0 1 1 2 2 3 3 4 4
0 1 1 2 2 3 3 4 4 5
These have zero correlation. The example is bad since you're probably thinking "that's just a delayed correlation", but weird data is weird.

>> No.12593040

>>12592979
>which need to predict a single output value.
Without knowing how this single value relates to the whole dataset, we can't help you.

>> No.12593054

>>12592696
If it's a question about a greatest common divisor, maybe you can divide something? Also, do you know how do for two numbers

>> No.12593110

>>12592979
principal components up the data, then fit the network on the highest two or three principal components

>> No.12593145

>>12592696
[eqn]a_1\mathbb{Z} +a_2\mathbb{Z}+...+a_n\mathbb{Z}\supseteq d\mathbb{Z}[/eqn]
is just Bézout's identity
[eqn]a_1\mathbb{Z} +a_2\mathbb{Z}+...+a_n\mathbb{Z}\subseteq d\mathbb{Z}[/eqn]
is a little more tricky but easy, you need to use the gcd's division property

>> No.12593302

in dimensional analysis, how do i know when to use ∝ instead of =. We've just started the topic

>> No.12593307

>>12592787
>is there a tidal componen
probably

>> No.12593309
File: 79 KB, 1561x531, tb.png [View same] [iqdb] [saucenao] [google]
12593309

>>12593040
Appreciate the hints and help.

It's a fixed length time series, with a small number of values per day in the series. Also has a ton of other global metrics that apply to the entire time series we're including too.

We're trying to predict one metric from the next day out of sample in the time series. For a concrete example:

We have 365 days of data for 10,000 shops. Each day includes data for:
- the number of toilet paper rolls sold per day
- the daily customers
- the shop turnover
- toothbrushes sold

This gives us a per-day vector of [math]\{ k_1, k_2, k_3... \} [/math]

Our global parameters would be something like:
- size of shop in square meters
- number of cash registers
- hours shop is open per day
- average tp rolls sold
- average daily customers
- std dev of tp rolls sold
- etc.

This gives us a global vector of [math]\{ g_1, g_2, g_3... \} [/math]

All of this data is concatenated into a single input vector, where [math]i[/math] is the day in question.

[math]
\{ k_{i0}, k_{i1}, k_{i2} ... k_{(i+1)0}, k_{(i+1)1}, k_{(i+1)2}, ... g_{1}, g_{2}, g_{4} ...\}
[/math]


We're trying to predict the number of toilet paper rolls sold on the 356th day for a given shop.

Hopefully this help clarify how retarded I am.

>>12593110
Thanks, reading up on this. I'm not sure how PCA would help if we're not seeing any correlation at all? At the moment we're seeing MAPE of like several thousand %.

>> No.12593421

>>12593309
>>12592979
There is something in mathematics and statistics known as the curse of dimensionality, which states that increasing the dimensionality of your feature space while keeping the same amount of sample points makes it increasingly harder to infer anything from your data.
Your feature space is 3500-dimensional, you won't infer anything of significance from it.

Trim off a significant portion of your variables, and you might see some interesting patterns.

>> No.12593495

>>12593309
>I'm not sure how PCA would help if we're not seeing any correlation at all?
Bro how tf are you charging for AI projects while having such little background? Use a method similar to pca regression for your inference problem

>>12593309
If the global parameters are constant, maybe make them a constant time series? Your concatenation approach is suspect as fuck and I would never do that.

>> No.12593545

>>12593309
Also damn I just realized you do not seem to be using matrices. Man are you trolling?

>> No.12593562

>>12593421
Thank you, anon. That makes a ton of sense. Will definitely try to eliminate as many variables as I can, and there's a whole ton we can get rid of.

>>12593495
Learning as I go, also charging the client peanuts for this as I'm far from experienced in this field.

You're right that the concatenation approach is suspect, but I couldn't find a way to train a single sequence model on multiple independent sequences, or even the terms used to describe that.

>>12593545
My statistics are extremely weak, although I've learned a lot more than when I first started this project. I guess at least I know that I'm pretty bad at it right now and doing what I can to learn. Thanks to you and others for the incredibly helpful responses so far.

How would matrices be applicable here?

At the moment I'm the dataset through a ton of different network topologies, different optimizers, activation functions and initialization strategies. Not expecting any good results from that based on what I've seen, so trying to figure out next steps from here.

>> No.12593581

>>12593562
>How would matrices be applicable here?
I usually help people here but you specifically said that you're charging money to an unsuspecting client for this, so I wont help you anymore, because your lack of basics is too ridiculous and the poor guy probably trusts you. You should be ashamed.

>> No.12593600

>>12593581
I can appreciate that, but it's part of a larger software project that I've been working on with the client and all other tasks and requirements have been more than successfully delivered on in the past 1.5 years.

Since we started this section, I've been completely upfront from the beginning about my lack of skills when it comes to this type of machine learning. I'm charging him less than 10% of what I normally would for the number of hours, as we're both taking it as an opportunity to learn together on this. I'm aware of my limitations and I fully intend to overcome them - I'm determined to learn enough to make this work, even if I have to be extremely humble and start from the beginning.

For the matrices, if you're referring to storing individual samples as rows and their associated input vectors as columns, I am definitely doing this.

>> No.12593650

>>12593600
Ok, it looks like you're not ripping him off at least. However, to be honest, the only way to help you significantly, given that you're in the dark, is to be in constant contact with you as a consultant. So good luck anon. Maybe look into time series analysis and discrete stochastic models to have an idea of what you're looking at, before throwing ML libraries at the data. Maybe use KNIME, which is free, and completely modular.

>> No.12593673
File: 3 KB, 625x400, mg iq.png [View same] [iqdb] [saucenao] [google]
12593673

Is this accurate or was the decay less consistent?

>> No.12593686

>>12593650
Thanks to you and all of the others in this thread.

Lots to learn. I'll definitely make a note and start going through those leads.

>> No.12593843

>>12592493
what the other anon failed to mention and is the correct answer to your question is that individual atoms do not have a preferential direction in free space. so these p-orbitals that are clearly non-spherical are not permanent, you can imagine the atom spinning around wildly such that the average electron cloud is still spherical on average.
in fact, permanent alignment of atoms/molecules without an external field is prohibited by symmetry.
>>12592714
do not listen to this anon's response, it is not only incredibly misleading but also flat-out wrong in certain aspects. you can absolutely have an atom with a highly nonspherical electron cloud, as long as you have it polarized by an external field.

>> No.12593850

>>12593673
>vertical axis

>> No.12593950

>>12593673
more like exponential decay

>> No.12593973

>>12592403
So, looking up the Dirichlet Green's function stuff to solve a problem (Laplacian of u equals 0 on the z>0 half of 3D space, u(r)=f(r) in z=0).
However, wherever I keep looking, like here: http://farside.ph.utexas.edu/teaching/jk1/lectures/node24.html
The Green function always comes up divided by -4*pi. Where does that constant come from?

>> No.12594030
File: 26 KB, 378x600, John L. Bell.jpg [View same] [iqdb] [saucenao] [google]
12594030

What am i in for?

>> No.12594034

>>12593950
I'd disagree since it feels like it was particularly harsh recently.

>> No.12594036
File: 29 KB, 448x260, 1610841581756.png [View same] [iqdb] [saucenao] [google]
12594036

"Find 'm' in F(x), if the main coefficient is equal to the independent term"
That's the translation

can someone please explain me why "3 cubed times 192" is the independent term?

>> No.12594037

>>12593973
IIRC unit sphere area.

>> No.12594041

>>12593843
>have an atom with a highly nonspherical electron cloud, as long as you have it polarized by an external field.
yes, that's what perturbation means
>you can imagine the atom spinning around wildly such that the average electron cloud is still spherical on average.
lolno.

>> No.12594052

Brainlet here.
In this paper:
https://arxiv.org/pdf/2101.05271.pdf
I have a hard time understanding Theorem 4.6 (pg 8) proof. Do we not care for the other paths since we just plug in for zero?

>> No.12594093
File: 15 KB, 712x259, Screenshot 2021-01-17 172405.png [View same] [iqdb] [saucenao] [google]
12594093

As more people get the Chinese Coof vaccine, is it a bad assumption to think that the "R" number for that state will go down by the % of people who have been vaccinated

So if a state's Chinese Coof "R" number is 1.05, and then they vaccinate 6% of their people with a 90% effective vaccine, should we expect "R" to drop (barely) below 1?

>> No.12594139

>>12594052
No, no, no. He's assuming he has an [math]arbitrary \ path \ A(t)[/math] and he's showing that the differential satisfies the condition (that is, is in the proposed Lie algebra).
He doesn't actually argue for existence of paths in every direction with [math]X + X^T = 0[/math], tho, so he loses one hundred points.

>> No.12594140

>>12594041
>lolno
??? you are wrong unless you choose to explain yourself.
free hydrogen atoms excited to a p state do not have a permanent quantization axis, meaning the electron cloud must be spherical on average

>> No.12594145

>>12594140
spherical in the laboratory frame*
to the hydrogen atom in a p-state, the electron cloud is the typical dumbbell shape. but to an outside observer the time-averaged electron cloud will be spherical, unless something is polarizing it

>> No.12594192

>>12594139
Gotcha, I think I understand what you are saying.

>> No.12594209

[math]\forall a, b \in \mathbb{Z} : a|b[/math]

is it this the correct negation?
[math]\exists a, b \in \mathbb{Z} : a\nmid b[/math]

>> No.12594320
File: 12 KB, 823x91, question.png [View same] [iqdb] [saucenao] [google]
12594320

I don't even understand the question. The author has defined inverses for "laws of composition", i.e. functions that take two elements of a set S and map them to S. That is, ab = ba = 1, being that b is the inverse of a. But here I'm given a function that just maps S->S, rather than SxS->S.

>> No.12594578

>>12594320
For the first part a hint is sort of in the way they posed the question. Assume there is an inverse, and then show that implies there is a map to 0 which becomes a contradiction since 0 is not in the set of natural numbers they used.

>> No.12594632

>>12594578
But why does that matter? There is no requirement for the inverse to map the natural numbers into the natural numbers. Or is it implied that the inverse of s, should again be a function from N->N?

>> No.12594638

>>12594320
If you have a function [math]f : X \rightarrow Y[/math], a left inverse is a function [math]g : Y \rightarrow X[/math] such that [math]g \circ f = id_X[/math], where [math]id_X[/math] is the identity function on [math]X[/math]. A right inverse is the same except [math]f \circ g = id_Y[/math].

>> No.12594670

>>12594632
The inverse of the composition is the identity. So in this case assume r is your right inverse, and then s(r(n)) = n for any n in your set of natural numbers N. So it is from N->N. Then suppose r(1) = m. Then if you look at s(m) = m+1=1 this implies m =0 which is not in N.

>> No.12594693

>>12594670
Yes, I figured that much. I just didn't know whether the inverse needed to operate on the same set as the function itself.

>> No.12594726
File: 208 KB, 684x735, 1610812858-20210116.png [View same] [iqdb] [saucenao] [google]
12594726

Scientifically speaking, how do you play unintentional spot the differences?
The left arm above and below obviously counts. Right hand's line and sofa creases are pretty distinctively different.
But do the ears count? The vertical line on his left hand?

>> No.12594768

I need help from people with experience working as engineers. Currently I going for a degree in Mechanical Engineering Technology. I chose this major simply because my school doesn't offer a full ME degree. I figured that the difference wasn't a big deal because I could fill in the gaps myself and pass the FE or PE exams to get licensed as an engineer. Now I am learning that in some states you can't even take those exams with an engineering technology degree, although luckily in my state it is accepted. My question is, how much will this affect my chances at becoming an true engineer with a high paying salary? Some companies might not care but now it seems like many will. I would switch schools right now without a thought, but I am almost 30 already and I have spent way too long in school. I'm sure not all of my credits will transfer and I will end up spend more time in school then I expected. My transcript is a mess even though I've maintained a good GPA. I'm not sure a better school will take me. Also, I would prefer to start working asap because I am sick of wasting my potential away in school. What do you think I should do?

>> No.12594785

>>12594768
Realistically you'll probably have to settle for lower-paying position straight out of school than if you'd have done a full ME degree. However, work experience is much more valuable than your degree. I'd recommend you stay, and finish your degree, and then get your license. Apply just as you would have if you'd done a regular ME degree. Either you get lucky and get a job where you want, i.e. would have gotten with an ME degree. Otherwise, take the lower-paying option and then when you have work experience just start applying to those places.

>> No.12594822

>>12594785
Ok so in most cases work experience should be enough to make up for the difference, that's what I was hoping for. Thanks for the advice.

>> No.12595129

What would be the equivalent of water for aliens living on cold planets like Mars or the moons of ourbgas giants? Ammonia?

>> No.12595543

>>12592403
What are the least set theoretic books on real analysis and basic abstract algebra? I can't just go into numerical analysis and computer algebra because they expect those as prerequisites, also some level of classical perspective of the topics is nice to have.

>> No.12595651

>>12595543
What are you trying to avoid in set theory? The real numbers are a set, measures are defined on a set, etc. It's hard to avoid the basics of it.

>> No.12595678

>>12595651
I'm not totally avoiding it, it's just that it's not a very useful intuition for computational needs. I'm not planning to go into pure math, but I want something that gives an unbiased (of it's computational issues) perspective on classical results or maybe gives a more computationally loaded alternative.
I've been checking out ASD papers and books by vickers, but it's a bit tough on the necessary background, seeing how it's still an active research area. The couple introductory books I've seen that do this are somewhat low-quality.

>> No.12595843
File: 24 KB, 200x300, skittles.jpg [View same] [iqdb] [saucenao] [google]
12595843

I bought pic related

is it going to be as fun as it looks (after I start and finish linear algebra by gilbert strang).

>> No.12595879
File: 408 KB, 579x800, __imaizumi_kagerou_touhou_drawn_by_wool_miwol__44deb6509063c236fc6611cb311a6f9a.png [View same] [iqdb] [saucenao] [google]
12595879

>>12595543
>>12595678
>numerical analysis
>actually caring about foundations or constructivist maths
It was only last semester that I was taking a grad class on the subject and having to read my professor's friends' recent papers, did things really change this fast?

>> No.12595889
File: 139 KB, 850x1511, __yorha_no_2_type_b_nier_series_and_nier_automata_drawn_by_gorgeous_mushroom__sample-c50e82462d151851d946931168192323.jpg [View same] [iqdb] [saucenao] [google]
12595889

Is it possible to create a circuit which does calculations that cannot be done in software if the dedicated hardware doesn't exist? Or perhaps is many (emphasis on many) orders of magnitude slower if the dedicated hardware doesn't exist?
I assume the answers are no (Turing completeness) and yes (coin mining algorithms for instance), but I'm not entirely sure.

I ask because I want to know whether I can make an algorithm that can 'only' run on a computer equipped with the dedicated hardware, even if people reverse engineer the source code (that is, the output of the hardware exclusive/accelerated operation must be included in the calculation of the result of the algorithm).

>> No.12595902

>>12595879
I didn't mean code monkey analysis, I meant computational analysis and the like.

>> No.12595927

>>12595902
I didn't mean code monkey analysis either, idiot. I meant a grad class about a specific subfield of numerical analysis taken in the maths department.

>> No.12595938

>>12595927
Then what relevance does it have to my question? You obv understand that's not the kind of analysis I'm talking about nigger

>> No.12595945

>>12595938
I'm just palpably confused by whatever the fuck it is you're calling "numerical analysis".

>> No.12595947

>>12595945
That's fine, don't worry about it

>> No.12595950

What are materials that hold multiple magnetic phases in its structure?

>> No.12595967

>>12595950
What do you mean by a magnetic phase in that context?

>> No.12595979

>>12595543
You can't completely avoid set theory for algebra since groups are defined on sets etc. Here's a good free book for algebra/number theory written by a computer scientist:
https://www.shoup.net/ntb/

>> No.12595983

>>12595967
Ferromagnetic-antiferromagnetic-ferrimagnetic

>> No.12595984

>>12595979
Thanks, exactly what I was looking for

>> No.12596001

>>12595984
Sorry I don't know anything similar for analysis. In my experience you just have to suffer through a standard intro to real analysis, then you can better learn from more specialized material. Not much of the material is directly applicable but experience with the style of proofs/definitions is helpful in understanding later material.

>> No.12596006

>>12596001
It's fine, algebra was the bigger hurdle. There's some literature out there that fits the bill for analysis, but requires some heavy algebra.

>> No.12596029

>>12595983
I'm now aware of any naturally occurring elements or minerals that fit. A paramagnetic material would have lots of magnetic domains but they would differ only in orientation, not ferro / anti-ferro properties, I believe.

I'm sure you could artificially create a layered material though. In fact I'm certain it has been done but I can't remember any references.

>> No.12596056

>>12596029
A magnetic 2D material?
https://en.wikipedia.org/wiki/Magnetic_2D_materials

>> No.12596090

>>12596056
Any layered 3D material could fit, 2D is a specific case with it's own special properties. It's an interesting area as it is possible high temperature spin-wave superconductors are related to these structures.

>> No.12596097

>>12596090
I see, thank you.

>> No.12596143

frens do any of you have a pdf or ebook copy of How to Prove It by Daniel Velleman?

>> No.12596155

>>12596143
libgen nigga

>> No.12596187

>>12596155
fuck me i'm a brainlet thanks nigga

>> No.12596276

In Rayleigh Scattering, when the incident radiation hits a scatter center does the scatterer absorb some/all of the incident energy and radiate it back out in all directions? e,g say we have white light scattered by molecules in the atmosphere, does the particles "absorb" most of the blue light and radiate it out spherically? The way I understand RS EM radiation excites charges to induce further EM radiation.

>> No.12596283
File: 156 KB, 900x900, __sekibanki_touhou_drawn_by_poronegi__1340bec645eaeb261e8a63cd77cb8293.jpg [View same] [iqdb] [saucenao] [google]
12596283

>>12595843
The /mg/ discussion over this was painful to read, so I just downloaded it and flipped over it.
It's pretty alright. No major complaints. Images are pretty. Good exercise density. Applications.

>> No.12596286

>>12594030
Infinitesimal Analysis

>> No.12596294

>>12596283
Nails on chalkboard, LMAO. I think someone might have some social deficiencies..

>> No.12596300

>>12592493
from a quantum mechanical point of view, and in my limited understanding of it. The symmetries in particles are finicky since the orbitals do not tangibly exist in a manner than can be analyzed as if they're permanent. Since the orbitals only show a probability density the electron can have any position at any time within the allowed energy shell. True symmetry then depends entirely on probability. That is to say unless youre not interested in the electrons themselves but only their probability densities.

>> No.12596321

>>12596283
Yeah I just had a look at it too, it's fine. It's a bit chatty and it doesn't cover much material in the abstract sense, but it has the applications. It's fine for an introduction to the subject, you can read something more in depth later. They seem to be acting elitist or something. It's not like it has bad information in it.

>> No.12596339

>>12596321
Holy shit I think I was having a conversation with you in the other thread.
I'm sorry: >>12596294
I shouldn't have said that. You taught me a lesson by rubbing it in. That has merit, at least. People shouldn't say things that might be offensive to vulnerable people.

>> No.12596350

>>12596294
Your post is hard to read and I actually can't tell if you're making fun of me or not, do you mind not talking sideways?

>> No.12596352

>>12596350
it is I:
>>12596339

>> No.12596364

>>12596352
That doesn't help at all, it just confuses me more.

>> No.12596371

>>12596276
I found this, https://www.alanzucconi.com/2017/10/10/atmospheric-scattering-3/ , if anyone else is interested in this. Also from the fermi-golden rule from condensed matter physics it can be reasoned that energy conservation is taking place. Therefore if some/all energy is absorbed then the same of energy is radiated in the direction given by the link above.

>> No.12596385

>>12596364
I'm sorry. You were acting like a psycho or high functioning autistic, to be fair.
I definitely knew what I bought was good before I bought it on amazon.
It's been several months since then, so I can't really comment on it because I haven't opened the book in a while.

>> No.12596775

should i try to do my masters thesis with some chinese professor that says he has gotten students into good phd or jobs that has interesting research but not exactly what i want to do, or go with a professor closer to my research interests?

>> No.12596810

>>12596775
a math qt that promises to have anal sex with you

>> No.12596822
File: 1.01 MB, 944x708, math is for gaylords, i&#039;d rather rape girls.png [View same] [iqdb] [saucenao] [google]
12596822

>>12596810

>> No.12596832

>>12596775
Just to make sure, the other professor is caucasian, right?

>> No.12596858

Hello poorfag from poor country in which the cost of food is steadily rising. Im working out in an improvised home gym while on a cut. 173cm 92 kg 32yo

I wanted to know if the following diet is acceptable for at least a couple of months:

Day A:

Breakfast: 1 banana 1 apple and one extra fruit
50ml of skim milk with coffe

Lunch: 2 boiled eggs 1 tblspoon of olive oil and assorted greens (ocasionally a different vegetable)

Dinner: 300-400g chicken breast

Free mustard, tomatoes and knorr broth

Calorie total: 1200-1500

Day B: exactly the same only 50g of rice instead of chicken breast

Alternate Day A and B. Will this be too harsh on my organism if done for a couple of months? what major nutrient am i missing

>> No.12596907

>>12596832
there are a few others i was looking at and most are caucasian, one is indian

>> No.12596912

>>12596775
this may surprise you but actually consider academic merit above race when deciding on your career

>> No.12596943

>>12596912
did you read the question or just have a very big nose

>> No.12596950

>>12596907
Right.
Like, I'm sure you're aware that all advice I'm going to be is essentially shitposting, right.
But you might want to look up his previous students to see if they were all chinese. Just to make sure you actually have a shot at borrowing his connections, yannow?

>> No.12596976

>>12596276
Rayleigh scattering is mostly elastic, meaning that almost all of the energy is re-radiated

>> No.12597052

>>12596858
i think you're looking for fit

>> No.12597081

any books on sacred geometry that don't go into new age spirit healing type shit and instead just focus on either than maths or the history of geometric worship?

>> No.12597087

>>12597081
Its called just maths bro.

>> No.12597157

>>12596912
>actually consider academic merit above race
if >90% of their students are foreigners of one race it means they're the ones not being impartial

>> No.12597347

Imagine your mother is looking for your PSP. She knows it's either in place A or place B, with probability [math]p[/math] of it being in place A.
If your PSP is in place A, searching in place A for [math]t[/math] seconds has a probability [math]\lambda_A t e^{- \lambda t}[/math] of her finding your PSP. Similarly for place B, with a [math]\lambda _B[/math] (these probabilities are wonky, just think Poisson process plus appropriate independence conditions).
If she's searching in one place, going to the other place to search there takes one second.

What's your mom's optimal strategy for finding your PSP the fastest?

>> No.12597465
File: 6 KB, 225x225, images.jpg [View same] [iqdb] [saucenao] [google]
12597465

I applied to a graduate program with a double edged sword recommendation letter. The professor doesn't know that I KNOW that he's a manipulative rat bastard. I would not be surprised if he wrote a bad letter to get me to stay working for him.

If I get rejected, can I contact admissions asking for specificities, and maybe a chance to defend myself?

>> No.12597619 [DELETED] 

Looooonely.
oka-tan#8719

>> No.12597727

>>12595889
> Is it possible to create a circuit which does calculations that cannot be done in software if the dedicated hardware doesn't exist?
No.
> Or perhaps is many (emphasis on many) orders of magnitude slower if the dedicated hardware doesn't exist?
In theory, you can make the speed ratio (dedicated hardware vs software) as high as you like. But you're up against the fact that the general-purpose CPU/GPU industry has huge investment and economies of scale. So a chip which can outperform a high-end GPGPU by several orders of magnitude might cost billions to design and have similar per-unit costs at scale, yet you probably don't have the same economies of scale as nvidia/AMD.

> (coin mining algorithms for instance)
The main benefit of ASICs for coin mining is energy efficiency. A 1W ASIC could do the same work as a GPU pulling 300W.

> I ask because I want to know whether I can make an algorithm that can 'only' run on a computer equipped with the dedicated hardware
The usual solution is to "hide" data (e.g. crypto keys) in the hardware. So while it would be possible to implement the calculation in software, it would require information which cannot feasibly be obtained even by someone in possession of the hardware.

>> No.12597883
File: 179 KB, 925x1480, aa460fd3f257f723c215e74085be1e678.jpg [View same] [iqdb] [saucenao] [google]
12597883

I got bored of this website, bye bye.

>> No.12598122

is the extended real number system not a field since there is no inverse for [math] \infty [/math]?

>> No.12598210

>>12598122
Yes. It's not a ring either. They don't form a group under addition.

>> No.12598232

Is Pinter really that bad? Some anon in the Pinter thread said it was for idiots or something.
I've seen it recommended in the latest math chart, and multiple times before by diff anons.

>> No.12598272
File: 17 KB, 725x545, beWCTKP.png [View same] [iqdb] [saucenao] [google]
12598272

how come noise is much more audible when i am walking on a hill and the person is either across on the other side or in the depression?

>> No.12598283

>>12598272
because there aren't any cars honking in traffic

>> No.12598292

>>12597883
anime bro leaving?? FUCK

>> No.12598669

i want to extract elemental mercury because it sounds like a nice project.
do i really need a fume hood if i don't want brain damage?

>> No.12598718

>>12598669
It's not a 'nice' project, it is a very stupid thing to do; since you are asking whether you need a fume hood or not implies you are below even what could be considered 'amateur lab work' or 'amateur chemistry'. It is basic lab safety equipment which is obviously essential for handling such a dangerous substance, depending on your method of extraction.

It would be much safer and much cheaper than buying expensive lab equipment if you were to enroll in a community college / night school / university on some chemistry course so you could learn more and use professional and expensive lab equipment that is maintained by trained technicians. I can't imagine why as such a total newbie to chemistry you would start with extracting elemental mercury, yet knowing it's toxicity.

>> No.12598727

>>12598669
nah go for it. just don't drink it!

>> No.12598818

>>12598718
i have enough chemistry experience (especially related to pyrometallurgy) but i have never worked with mercury before and i don't have access to fume hoods because of lock-down. i could distill it in my backyard but i don't want my neighbors to get butthurt so i'm thinking about leaching it but i don't know if a hood is strictly necessary or if i can just let it sit outside.
your post sounds like midwit undergrad projection and you didn't even answer my question, apply yourself faggot.

>>12598727
thanks anon.

>> No.12598976

>>12598818

Asking whether or not you need a fume hood strongly implies you know basically nothing about what you're doing. It is a basic piece of equipment that you'd know you need if you knew how to extract the substance. How about you apply yourself you incompetent fool, I don't need to answer such a stupid question.

And, I'd recommend not extracting mercury outside. It's a nuisance to your neighbours and it'll probably be against local environmental regulations - provided you don't live in a third world country.

>> No.12599141

How do I solve for [eqn]166642=\frac{1}{8}\left(x^2-x+600\left(\frac{2^{\frac{x}{7}}-2^{\frac{1}{7}}}{2^{\frac{1}{7}}-1}\right)\right)[/eqn]?

>> No.12599182

Can someone explain to me what's the difference between these two formulas below for standard deviation and when to use one or the other?

[eqn]
\sigma_x = \sqrt{\frac{1}{N}\sum_{i=1}^{N} \left(x_i - \bar{x} \right)}
[/eqn]

and

[eqn]
\sigma_x = \sqrt{\frac{1}{N-1}\sum_{i=1}^{N} \left(x_i - \bar{x} \right)}
[/eqn]

where [math]\bar{x} = \frac{1}{N}\sum_{i=1}^{N} x_i[/math] is the arithmetic mean.

>> No.12599198

>>12599182
one has N-1 and the other has N

>> No.12599218

>>12599182
>https://en.wikipedia.org/wiki/Bessel%27s_correction

>> No.12599221

>>12596858
https://tdeecalculator.net/

In my opinion you should spread out the calories throughout the day so that you don't feel miserable. Also, take some multivitamins and fish oil. I would do it this way instead

>Breakfast
1 banana 1 apple and one extra fruit
50ml of skim milk with coffe
2 boiled eggs

>Lunch
150-200g chicken breast + veggies/salad

>Dinner
150-200g chicken breast + veggies/salad

Start with that and see how you feel, then adjust the macros accordingly, at the end of the day it all comes down to calories in/out. Since you're working out you'll definitely be craving carbs all the time. Carbs aren't your enemy, you just have to keep track of your calories, for example, a bowl of oatmeal an hour before working out would be great.

>> No.12599228

>>12599221
By the way, I used your own meal plan as an example, I personally would switch the fruits for oatmeal, so something like this

50g oats with 100 ml water oatmeal (just microwave it for 1.5-2 min), add some cinnamon and sugar substitutes
50ml of skim milk with coffee
2 boiled eggs

>> No.12599307

>>12599218
So basically I only have to use this correction if I'm working with samples of a much larger population?

>> No.12599456

>>12599141
Numerically. You can't solve a^x=x^b (or anything similar) algebraically.

>> No.12600283

>>12599307
No the bessel correction is needed to make the variance estimator unbiased.

>> No.12600309

What's the motivation behind Polish spaces, /sci/? What interesting things can be said about a completely metrizable space, when the metric itself has been forgotten? One can't even speak of Cauchy sequences in this setting, so it all seems rather... pointless.

>> No.12600811

Is there a way to relate Whispering Gallery Modes (wave propagation around curves) to linear modes (simple harmonics). Can i just stretch the curve straight without loss of generality?

>> No.12600968
File: 14 KB, 499x257, bar-magnet-500x500.jpg [View same] [iqdb] [saucenao] [google]
12600968

If you break a magnet in half, do each of the two halves acquire their own south and north?

>> No.12601057

>>12600968
yes

>> No.12601111

>>12600968
south and north, as the name implies, are directions.
magnetism is a direction.
What is north without a south?

>> No.12601120

>>12598818
>your post sounds like midwit undergrad projection
frankly, so did yours.
outside should be fine. Don't breathe the vapors, and don't do it more than once or twice.

>> No.12601255

>>12600968
Think if it like this if you have two arrows in a row: ->-> and break those in half you still have two arrows: -> and ->. Poles and magnetism works kind of the same way.

>> No.12602043

What’s the easiest way to make H2S gas with household chemicals and achieve 1000ppm? Should be pretty obvious why. I appreciate anyone that can answer.

>> No.12602205

I'm struggling to complete my first linear ODE problem.

Given [math]y' + \frac{2xy}{x^2+4} = \frac{3x^2}{x^2+4}[/math], with an initial value of y(0)=2, I proceed as:
[math]y' + Py = Q \rightarrow P = \frac{2x}{x^2+4}, Q=\frac{3x^2}{x^2+4}[/math]
[math]\rho = \exp(\int P dx) = \exp(\int \frac{2x}{x^2+4}dx) = exp(\ln(x^2+4)) = x^2+4[/math], and
[math]\rho'=2x[/math]

Thus
[math]\rho'y + \rho y' = (py)' = Q \rho = \frac{3x^2}{x^2+4} \cdot (x^2+4) = 3x^2[/math], and
[math]\int(\rho y)'dx = \int 3x^2 dx = x^3 + C [/math]
Then [math] y = \frac{x^3 + C}{\rho} = \frac{x^3+C}{x^2+4} [/math]

So [math] y(0) = 2 \rightarrow C = 8 [/math]


phew.... did I do all that right?

>> No.12602306
File: 546 KB, 1235x698, Giraffe.png [View same] [iqdb] [saucenao] [google]
12602306

I asked a homework question and you all called me dumb for getting it wrong. Well it turns out there was a problem with the stupid fucking online homework system and I was right the whole time so fuck you all.

>> No.12602332

>>12602306
lol what was the question?

>> No.12602402

>>12602306
i believed in you the whole time anon c:

>> No.12602446

>>12602205
yes

>> No.12602496

>>12602446
thx

>> No.12602645

>>12602332
I asked how to tie my shoes, but /sci/ couldn't figure it out.

>> No.12602664

[math] \forall a,b \in S : aRb : \iff 3|(a-b) \text{ or } 2| (a-b) [/math]
can someone explain what this equivalence relation even means? specifically i don't understand what 3|(a-b) and 2|(a-b) means. to me it reads 3 such that (a-b) which doesn't mean anything to me.

>> No.12602718 [DELETED] 

>>12602664
It means 3 divides a-b and 2 divides a-b
[eqn]a|b

>> No.12602723

>>12602664
It means 3 divides a-b and 2 divides a-b

>> No.12602739
File: 10 KB, 712x87, bigly.png [View same] [iqdb] [saucenao] [google]
12602739

is this right?

[math]\int \frac{x}{x(x-1)}dx = \int 1 dt = t+c [/math]
[math]\int \frac{-1}{x}+\frac{1}{x-1}dx = \ln(x)+\ln(x-1) =\int 1 dt = t+c [/math]

natural log is undefined at ln(0), so I think I did something wrong as x(0) = 3 couldn't happen then right?

>> No.12602745
File: 1.31 MB, 1920x1080, 5LYzTBVoS196gvYvw3zjwLSrjfQw1ybyyl8LBPcPxyc.png [View same] [iqdb] [saucenao] [google]
12602745

>>12592403
what makes a mathematical theorem "interesting"? this seems like a fundamental problem but has not been seriously touched

>> No.12602819

>>12602739
It's [math]t[/math] that is worth 0 at the initial value so it's not a problem, keep going and you'll see you can get a solution. Also beware, you should be having [math]\ln{(\frac{1}{x})}[/math] on your LHS, not [math]\ln{(x)}[/math].

>> No.12602849

>>12602819
>[math]\ln(\frac{1}{x}[/math]
how so?

>> No.12602869

>>12602849
[math]\frac{-1}{x}[/math] integrates as [math]-\ln{(x)}[/math] (or [math]\ln{(\frac{1}{x})}[/math], same thing).

>> No.12602882

>>12602869
oh shit yeah, forgot -ln(x) was equal to that and did notice I omitted the negative in the original post, meant to clarify. so I still don't really get it though, this is only my second or so noteworthy problem in diff eq so sorry for being underread. but i have

[math]-ln(x) + ln(x-1) = t+c [/math]

and you say that x(0) just means t is worth 0. so -ln(x) + ln(x-1) = 0 + c = c... ? i just don't really get what I'm looking at here.

>> No.12602915

>>12602882
[math]\ln{(x(0)-1)} - \ln{(x(0))} = c[/math]

>> No.12602932

>>12602915
fck I should've got that, I guess i wasn't really appreciating that x was a function.. thanks

so to be clear the answer to the question is jsut the value of C, rihgt? given numerically ofc

>> No.12602941

>>12602932
I'd say the answer is also the function [math]x(t)[/math]

>> No.12602976

>>12602941
thanks for the help, not sure what my deal is with this stuff. it's not that it's hard, it just feels more like I'm directionless when solving the problems.

>> No.12603015
File: 21 KB, 773x157, ezpz.png [View same] [iqdb] [saucenao] [google]
12603015

so this one just seemed too ez, the answer is just x(t) = 0, -3 or 3. boom, roasted. is that all? other than the phase diagram requirement

>> No.12603023

^ with b) just being x(t) = +/- 1. just seems stupid easy I must be misunderstanding

>> No.12603077

>>12602976
i recommend this site:
https://tutorial.math.lamar.edu/Classes/DE/DE.aspx

also check your solutions on the net

>> No.12603085

>>12602723
I had no idea | could mean divide. So then it isn't an equivalence relation for the integers, right?

>> No.12603471

>>12603085
If a divides b, is it necessarily true that b divides a?
Transitivity does hold, though, and obviously so does reflexivity.

>> No.12604010

Would it be safe to take vyvanse pills that expired almost a year ago? Probably not taking anyway but I think it might be one of those cases the expire date is there just to make you waste money.

>> No.12604187

>>12597883
post tits b4 you anhero

>> No.12604199

>>12604010
Expiry dates reflect the fact that drugs degrade over time. An expired drug may cease to have its intended effect.

>> No.12604369
File: 44 KB, 225x320, 211230.jpg [View same] [iqdb] [saucenao] [google]
12604369

I wrote a physics-based isekai
I hope you guys like it
I really enjoyed writing this

>https://docs.google.com/document/d/e/2PACX-1vTYXTCND_nQbbRQ5KttQ4ZJ-Jf3E_Z7c40alH8utYtRJO-RrYk07tn-0HEM7-fCtsDBpL6Z9LEU0TtX/pub

>> No.12604382

Please explain how quantum is related to neural networks.

>> No.12604413

>>12604382
It's not related in any way.

>> No.12604451

Struggling to construct a counterexample that shows that the projection of a measurable set in the product space need not be measurable in the component space. Apparently there exists a Borel set of [math]\mathbb{R}^2[/math] whose projection is not a Borel set of [math]\mathbb{R}[/math], but I'd like to find a simpler example. Any ideas?

>> No.12604466

>>12604413
Neural networks are spin glass models which are quantum systems. So shush.

>> No.12604491

>>12604382
Neural networks are a clever math trick (from analysis).
Quantum mechanics is a working description of physical phenomena.

Completely different imho

>> No.12604494

>>12604491
>what is a restricted Boltzmann machine?

>> No.12604516

>>12604494
1. The theoretical legitimacy of ANNs comes exclusively from functional analysis and approximation theory, with no connection to any physical phenomena.
2. If a physical model turns out to be structurally equivalent to a neural network, it does not mean that neural networks in general have any physical legitimacy per se.
3. Midwits like you should leave the board

>> No.12604653

In optics, light is defined as an electromagnetic field, a wave that is defined everywhere in space. In everyday life (and also in refraction experiments), we think of light as thin rays. How do I understand light rays in this context? How do rays decay in such a way that they still satisfy Maxwell's equations?

>> No.12604672

>>12604653
Optics in terms of rays is called 'Geometric Optics', it's an abstraction of how it really works. It ignores diffraction and interference and so the rays don't decay in this picture of things.

>> No.12604814

I'm interested in Genetics and gene editing.
What would you recommend would be the best gateway for it? I'm in the military so i'll be knocking out classes slowly but won't be getting that much hands on experience

>> No.12604830

>>12604814
Well, what background do you have in biology so far? For a start you'd want to specialize in Molecular Biology.

>> No.12604833

recommendations on nootropics?
I recently started smoking to help me think through things but that'll kill my lungs sooner or later. What do you take anons?

>> No.12604839

>>12604830
Not much took microbio in Highschool but that's it. My job in the military is a avionics mechanic but ill be able to reclass in a year.

>> No.12604868

Everyone knows that [math]x^2 − x + 41 = x(x - 1) + 41[/math] gives prime numbers for [math]x = 1, 2, …, 40[/math]. Since [math]x^2 − x + 41 = x(x - 1) + 41[/math], it gives composite numbers when [math]41 | x[/math] or [math]41 | x - 1[/math], so [math]x^2 − x + 41[/math] is composite when [math]x = 41, 42, 82, 83[/math] etc. Thus they are trivial composites. Some sources list 41 and 42 as counterexamples, so in a sense, they are trivial counterexamples. So my question is, are there any non-trivial composites/counterexamples for any positive integers of [math]x[/math]?

>> No.12604878

>>12604839
Well then you'd need to start with basic biology courses like a biology major would. I don't know how the military works but medicine probably isn't the best field since it'd be more like first aid stuff, you'd want anything to do with biology or chemistry labs.

>> No.12604923

>>12604672
What I mean is that I'm trying to see how the amplitude should decay in such manner that the electric wave obeys Maxwell's wave equation. This doesn't seem trivial.

>> No.12605142

>>12604653
>>12604923
What about light rays decay?

Maxwell is a more fundamental theory which says light is a wave. Geometric optics is an abstraction and is only valid for a more limited regime.

>> No.12605214

>>12605142
Let me put it in another way: if I shine a light with a laser, I create a thin, localized field propagating at the speed of light, whcih we call a beam. How does Maxwell's wave theory describe the dynamics of this process? Whenever I read about electromagnetism, light is presented as a vector field of the form [math]\vec{E} = E_0 \exp{(i k_z z - \omega t)}[/math], which defines a field over all of space rather than the beam I see in everyday life. How does this work?

>> No.12605219

>>12605214
I forgot the [math]\hat{x}[/math] unit vector there.

>> No.12605306

>>12605214
A laser beam is not something described by Maxwell's equations, it's not classical. The best you can do are plane wave solutions which are propagated in one direction.

>> No.12605318

>>12605214
That is the plane-wave solution for an EM wave moving only in the z direction with initial conditions [math]E_x^0, E_y^0[/math]. Since you didn't create the laser to start with over all space it won't then propagate over all space, it stays moving only along the z axis.

>>12605306
To create a laser does require quantum mechanics but once the light is emitted it can be treat like any other electro-magnetic wave.

>> No.12605332

>>12605318
Okay, clearly the plane wave solution does not describe the situation at hand. But then what does? If I think of a wave moving through a medium, I immediately imagine radial propagation, so why can lasers create a sharp, localized beam? I suppose what I'm asking for in other words is: how is the particle theory of light supported by Maxwell's equations?

>> No.12605383

>>12605332
The plane wave solution is only valid in a vacuum. Once it travels through a medium then dispersion does occur. Since Maxwell are second order differential equations and depending on the system there may not even be analytic solutions. So either you solve numerically or you make approximations and simplifications and you get back the less general optics theory.

The particle theory of light is not explained by Maxwell which is why they had to come up with Quantum Mechanics in the first place. Even that was later found not to be good enough which is why Quantum Field Theory and specifically Quantum Electro Dynamics are a thing. Lasers are due a quantum mechanical process to create coherent / correlated beams which ensures it is all emitted along a single direction.

>> No.12605399

>>12605383
So if I understand this correctly, one cannot explain laser beams by just using Maxwell's wave theory, and such a beam is not a stable solution to Maxwell's equations.

>> No.12605413

>>12605399
Maxwell applies to the beam just fine but to explain the laser in its entirety, as in how the light is created so precisely, needs quantum mechanics.

>> No.12605441

>>12604369
Where's the pysics?

>> No.12605449

>>12605413
>as in how the light is created so precisely
If you are referring to the atomic process that generates the emission, that is not the side of the issue that I am interested in at the moment. What I want to know is what solution to Maxwell's equations describes the propagation of a light beam, i.e. a wave that has a highly localized amplitude along its path of propagation and extends at the speed of light. After all, common sense would say that such a wave would disperse so I really would like a mathematical solution in order to be convinced that Maxwell works.

>> No.12605575

why isn't [math] f : \mathbb{R} \rightarrow \mathbb{R}; a \mapsto | \sqrt{a} | [/math] well defined? i thought it was because anyone absolute value of the square root of a would just return a postive real, right?

>> No.12605588

>>12605575
It's well defined and symmetric around 0

>> No.12605596
File: 34 KB, 468x489, Screenshot 2021-01-20 164913.jpg [View same] [iqdb] [saucenao] [google]
12605596

>>12605588
this was part of a multiple choice thing and I got it wrong.

>> No.12605606

>>12605596
Square root is undefined for negative numbers.

>> No.12605607

>>12605575
switch the order of abs and sqrt

>> No.12605621

>>12604382
what if
now
what if the inner quantum j is related to the hidden layer of a neural network?

what if
the Hund's rule of finding minimums and maximums is just another way of optimizing the weights?

what if
the tanh(x) function is both the activation function and the magnetization of linear materials?

totally a coincidence.

>> No.12605657

>>12605606
>>12605607
thanks, i indeed read it as [math] \sqrt{|a|} [/math]

>> No.12605729

>>12605596
What is wrong with 2. and 4.?

>> No.12605797

>>12605449
>common sense would say
by your logic, common sense might say that the heat equation works for arbitrarily complicated geometries. while this is true, we don't have any way of reaching an analytical solution.
to describe the propagation of laser light you can use maxwell's equations, but good luck finding an analytic solution.

>> No.12605826

>>12605575
Its derivative has a jump discontinuity at zero

>> No.12605834

>>12604833
Stick with caffeine

>> No.12605855

>>12602043
>should be pretty obvious why
You want to kys, and destroy the neighbors' property values?

>> No.12605927

>>12605797
>common sense might say that the heat equation works for arbitrarily complicated geometries
That's a total non-sequitur. Of course the heat equation provides solutions for any geometry; the question is whether the differential equation is capable of reproducing phenomena as basic as a simple light ray. The question isn't even that complicated, in fact it's rather concise. It reminded me of solitons and Green's functions, which both describe some sort of travelling wave and have exact mathematical expressions. And as far as my experience goes, such distinctive phenomena always have an elegant explanation, so with all that in mind I doubt that this problem is indeed too obscure to analyze.

>> No.12606636

Hey /sci/ I'm a CS brainlet and decided that I want to learn more math. As a part of my CS degree I had to take Calculus I-III and linear algebra. What is the next step for me?

>> No.12606754

If you were traveling at the speed light would you see everything else frozen in place because you dont experience time ?

>> No.12606848

>>12606754
You would see everything frozen because the faster you observe something moving relative to you the slower their 'clocks' are seen to run. So at lightspeed their clocks stop. Time still runs normally for the observer though.

>> No.12606868

>>12606848
if time stops wouldnt everything be instant for the one going at light speed ? but that can't be because we know light takes time to travel

>> No.12606876

>>12606868
Time doesn't stop for the observer no matter what speed they travel at, even the speed of light.

>> No.12606881

>>12606876
im not talking about the observer im talking about the theoretical person traveling at light speed

>> No.12606882

>>12606636
Proofs
Relearn Linear Algebra with proofs
Relearn Calculus 1-3 with proofs
Relearn Algebra with proofs

>> No.12606892

>>12606881
That is who I meant by the observer.

>> No.12606896

>>12606882
>Relearn Calculus with proofs
Lol no
It should be
>Proofs
>Real Analysis +Linear/ Abstract Algebra

>> No.12606900

>>12606892
time doesnt stop but they dont experience time. so why wouldnt they perceive there travel as instant if they were traveling to lets say a star 8 light years away

>> No.12606914

>>12606900
Time doesn't change for the light speed person but remember that time isn't the only thing that is relative. Length contraction would occur so the distant to the star would shrink to zero.

>> No.12606921

>>12606914
>Length contraction would occur so the distant to the star would shrink to zero.

so it would be perceived as instant for them then

>> No.12606927

>>12606921
Yes but not because time has stopped for them.

>> No.12606957

>>12606896
That's what I meant, the joke is he's going backwards with proofs.

>> No.12606993
File: 186 KB, 570x570, You should be able to.png [View same] [iqdb] [saucenao] [google]
12606993

>>12606636

Math for CS:
https://stackoverflow.com/questions/14868480/math-for-computer-science

https://4chan-science.fandom.com/wiki/Mathematics

There's also a thread for this stuff:
>>>/sci/cspg
>>>/g/cspg

>> No.12607168
File: 35 KB, 700x394, China coof vaccine.jpg [View same] [iqdb] [saucenao] [google]
12607168

Would it hurt the US's own vaccine efforts, or Moderna itself, to just release their "recipe" to the world so the shithole countries can make it? Or does the synthesis/scaling stuff involve a ton of trade secrets?

>> No.12607179

>>12607168
We know the recipe and the basics of the process. We don't have the technology or specific equipment that Moderna has to develop the vaccine on our own.

>> No.12607198

1. Z⊂S
2. ∀s∈S P(s)
3. ∀z∈Z P(z)
is there a rule rule of inferencethat states 3 follows from 1 and 2? i understand this can be proven with implications instead of quantifiers, but i feel like there should be a rule of inference for this

>> No.12607208

>>12607198
*a rule of inference that states

>> No.12607213

>>12607198
If there isn't a rule, make it one and see what happens

>> No.12607257

>>12607179
What about some country like India just handing them a cool billion dollars for it?

Or Russia just hacking them?

>> No.12607259
File: 29 KB, 640x480, Attempting to reason with PCs.jpg [View same] [iqdb] [saucenao] [google]
12607259

What questions can I ask at an upcoming quantum computing seminar so I don't look like a brainlet?

>t. brainlet HR person forced to attend

>> No.12607266

>>12607179
Also I'm not talking about reverse-engineering it, I'm talking about taking a known molecule and making it at scale.

>> No.12607268

>>12607259
Keep your fucking mouth shut.

>> No.12607278

>>12607213
all it says is what is true for all elements of a set is true for all elements of a subset of that set. i'd be surprised if it doesn't already have a name, even if only an informal one.

>> No.12607650

>>12603015
>>12603023
bump, just trying to check my answers here

>> No.12607675

>>12607259
how much quantum wood could a quantum wood chuck chuck if a quantum wood chuck chuck could chuck quantum wood?

>> No.12607714
File: 26 KB, 747x112, bunnyisland.png [View same] [iqdb] [saucenao] [google]
12607714

i am absolutely lost on this one

>> No.12607733

>>12607714
Find [math]P(t)[/math], use the fact that [math]P(t=0) = 1[/math] to determine the integration constant, then use [math]P(t=10\text{y}) = 5[/math] to determine [math]k[/math]. Then just calculate [math]P(t=20\text{y})[/math].

>> No.12607852

>>12607733
nice get, how do I find p(t) though? I'm sure I can get it once I get the ball rolling, but i'm having trouble getting there.

>> No.12607865

>>12607852
Same method as always.
[eqn]\int \frac{dP}{P(20-P)} = \int kdt[/eqn]

>> No.12607874

>>12607852
integrate dP/dt

>> No.12607883

>>12607865
that's exactly where I'm stuck lol

FUCK

oh wait I get it now, duh. still fuck though

>> No.12607888

>>12607874
tha'ts not even what I'm supposed to you ingrate

>> No.12607927

It's been some years since I've studied analysis (with Rudin's undergraduate book). How much of that material do I need to remember? Can I just jump into a graduate analysis book and be fine?

>> No.12608006

>>12607865
does this look right?

[math]\int \frac{dP}{P(20-P)} = \int kdt[/math]
[math]\ln(\frac{P}{P-20}) = 20(k\cdot t+c)[/math]
[math]\frac{P}{P-20} = e^{20(k\cdot t + c)} [/math]
[math] 1 +\frac{20}{P-20} = e^{20(k\cdot t + c)} [/math]
[math] P = \frac{20}{e^{20(k\cdot t + c)} - 1} + 20[/math]
[math] P(0) = 1 = \frac{20}{e^{20(c)} - 1} + 20 [/math]
[math] C \approx 1.5707[/math]
because it feels very wrong. so that c is my integration constant?

>> No.12608011

>>12608006
* c = 0.15707

>> No.12608025

the math above resulted in a final answer of about 20k bunnies, which seems reasonable but my process felt wrong.

>> No.12608081

>>12607268
>>12607675
Alternatively, what if I ask about engaging policymakers or some kind of economic application question?

>> No.12608309

>>12608006
You've got a sign wrong. log(P/(P-20)) will be undefined for P<20. You should get P/(20-P)=Ce^20kt. t=0,P=1 => C=1/19, t=1,P=5 => k=ln(19/3)/200. At t=20, P=95/7 ~= 13.571.

>> No.12608389

>>12607259
What do quantum and neural networks have in common?

>> No.12608415
File: 260 KB, 2000x1600, 2048.jpg [View same] [iqdb] [saucenao] [google]
12608415

>>12608389
Good god don't ask this.
>>12607259
Broadly speaking, there are no "talking points" we could possibly give you without further context to prevent you from looking like a brainlet. Put another way, asking a good question that indicates you are knowledgeable about the topic of the seminar requires you to understand the specific subtopics that were discussed. We could give you some random, legitimate questions to ask, but if they were essentially orthogonal to the context of the seminar itself, it would just seem like you were pulling shit out of your ass and using buzzwords that you personally don't understand.
Which would be essentially true, hence another reason not to do this. Just be honest and either ask a legitimate question that forms during your time there, or if none arises, don't ask anything. The age-old adage holds well here.

>> No.12608571

>>12607278
You do realize that naming every single inference rule that every idiot can think of in every possible language is just stupid, right? Maybe some dude calls whatever rule you're thinking about some name, but in practice, why would anyone care?

Anyways, you can get 3 by using a mix of instantiation, generalization, and modus ponens.

>> No.12608666
File: 841 KB, 600x600, 1611159227102.png [View same] [iqdb] [saucenao] [google]
12608666

So, I'm studying digital image processing algorithms.
I'm trying to decide which algorithms to use for which cases.
Say I have the following cases:
-Parking lot and I'm trying to find out how many cars are there (with as minimum amount of cameras as possible)
-Photo of an avocado, trying to infer wether it's ripe enough or not.
-Find out if there is a TV or not in a picture
Would you mind pointing me in the right direction? Is there anyone with experience in DIP?

>> No.12609057

>>12596155
Are there other sites for books other than lib since I can't find some Oxford primers

>> No.12609073

>>12609057
z-lib

>> No.12609311
File: 4 KB, 734x725, osc.png [View same] [iqdb] [saucenao] [google]
12609311

Could some EE brainchad bully me a bit for this circuit? I am just a dumb code monkey trying to get into analog electronics, and I'm scared that if I actually build it, I'll set my flat on fire.
It's supposed to be a voltage-controlled relaxation oscillator, if that's not obvious.
Haven't figured out how to find the correct values for the elements yet. Any tips appreciated.

>> No.12609313

>>12608666
All of these can be unironically approached using artificial neural networks.

>> No.12609342

>>12608415
SAY IT WITH ME
QUANTUM
EQUALS
NEURAL
NETWORK

>> No.12609542

Is wolfram pro worth it?

>> No.12609691

>>12609311
did you copy a working diagram?

>> No.12609776
File: 14 KB, 290x400, 308979.image0.jpg [View same] [iqdb] [saucenao] [google]
12609776

>>12609691
It's a modified version of picrel

>> No.12609995

>>12609542
You can get Mathematica instead and access Wolfram from there.
You can download Mathematica either legally or ...

>> No.12610020

>>12609311
>>12609776
Well for one all the current from 2 and 3 flows through the potentiometer, R6, straight to ground.

>> No.12610103

>>12610020
Oh, that makes sense. Could I add a diode behind R6 to prevent this?

>> No.12610122
File: 4 KB, 824x800, osc.png [View same] [iqdb] [saucenao] [google]
12610122

>>12610103
...like so?

>> No.12610239
File: 113 KB, 1575x1538, 2021-01-21 21_32_16-_sci_ - _sqt_ _qtddtot_ - stupid questions thread - Science &amp; Math - 4chan.png [View same] [iqdb] [saucenao] [google]
12610239

>>12610122
that diode won't do anything
try building your circuit in
https://www.falstad.com/circuit/

>> No.12610434

>>12610239
I build it and got it working with that modification plus some other ones. Thanks for you help

>> No.12610530

>>12608571
>in practice, why would anyone care?
people seem to get it wrong often. for example, not to be political, some people think the statement black lives matter implies not all lives matter. regardless of which side is correct politically, that particular misunderstanding is illogical.
>Anyways, you can get 3 by using a mix of instantiation, generalization, and modus ponens.
thank you

>> No.12611023
File: 15 KB, 681x274, Screenshot_20210121_225628.png [View same] [iqdb] [saucenao] [google]
12611023

Why won't sagemath simplify [math] cos(x)^2 + sin(x)^2 [/math] to 1?

>> No.12611032

I have a hard lymph node in my neck ,right side close to my jaw.I know I have to go to the doctor but what do you think it could be?
I keep reading cancer this cancer that and I'm shitting bricks.

>> No.12611076

>>12611032
could be cancer, more likely to just be an infection

>> No.12611296

>>12611023
because it's only true for [math]x \in \mathbb{R}[/math]
if you don't specify that constraint, it's not going to assume it

>> No.12611404

How would find the uncertainty of [math]2\pi \sqrt{\frac{L}{g}}[/math], where L has an uncertainty of [math]\pm 0.05[/math]?

Take L as 0.7 if you'd like, as this is the value I'm working with.

>> No.12611413

>>12609073
It seems like it isn't in there are there any alternatives like one I am looking for is d-block chemistry Oxford primer

>> No.12611869

Was HIV legitimately proven to be a/the probable cause of AIDS?

>> No.12611885

>>12611404
https://en.wikipedia.org/wiki/Propagation_of_uncertainty#Simplification

assuming 0 uncertainty in g, uncertainty of [math] f(L)=2 \pi \sqrt{\frac{L}{g}} \rightarrow \sigma_f = \sqrt{ (\frac{ 2 \pi }{\sqrt{g}} \frac{d}{dL} \sqrt{ L })^2 \sigma_L^2 } [/math]

hopefully you can take it from here

>> No.12612088

>>12611869
Well most people think so, but you can read about it here for a general overview of why they think so:
https://www.nejm.org/doi/full/10.1056/nejmp038194

>> No.12612661

Whats the deal with dreams ? I had one last night and felt much better than other nights

>> No.12612827
File: 17 KB, 563x244, Screenshot_20210122_095643.png [View same] [iqdb] [saucenao] [google]
12612827

>>12611296
It's true for [math] x \in \mathbb{C} [/math] too isn't it? There may be some other sets on which sin and cos can be meaningfully defined, in which it isn't true, but I doubt it. Regardless, sage still won't simplify given this information. I've realize that there's a couple of other functions: simplify_full and simplify_trig which actually do have the intended effect though.

>> No.12612830

>>12612661
Did you check your pants in the morning?

>> No.12612927 [DELETED] 

>>12608006
[eqn] \int \frac{dP}{P(20-P)} \neq \ln\left(\frac{P}{P-20}\right) = 20(k\cdot t+c) [/eqn]
you need to use partial fraction decomposition to do that integral

>> No.12613032

>>12612830
No wet dream
I only remember fighting off criminals and I entered some kind of machine and was amazed by its capabilities and complexity

>> No.12613119
File: 442 KB, 3464x3464, 1587319733447.jpg [View same] [iqdb] [saucenao] [google]
12613119

Velocity:

v = u + at

Displacement:

s = vt

Substitute first eq in second eq:

s = (u + at)t
s = ut + at^2

But this is wrong, it should be

s = ut + 0.5at^2

Explain?

>> No.12613124

>>12613032
Describe it

>> No.12613128

>>12613119
this board is not for high schoolers

>> No.12613136

>>12613119
I remember when I didn’t know about calculus..

>> No.12613142

>>12613128
>nooo you can't try to teach yourself stuff when you're older!
It looks like you're the child.

>>12613136
Another one concerned more with looking smart than helping others, in a thread about dumb questions.

lmao fucking useless

>> No.12613172

>>12613119
>>12613142
Because that is not the correct formula for the displacement. s=vt is only true if the velocity is constant. In this case it's not since there is acceleration present.

>> No.12613174

>>12613172
Ah thanks that's the detail I overlooked.

>> No.12613180

>>12613119
s = vt is for constant v only
>>12613136
it has nothing to do with calculus retard

>> No.12613273

>>12613124
I already forgot Most of it
Happens with dreams all the time

>> No.12613450

>>12613119
https://www.youtube.com/watch?v=ZJTtyYKKsuI

>> No.12613639

How to show that the exponential function (series representation) is uniformly convergent for every complex number?

>> No.12613714

can someone explain what a tensor (field) is in a way a 6 year old/engineer can understand? I think I understand a specific application of them, but I can't generalize it and "see the big picture"

what I think I understand is the stress tensor. I'm an EE so sorry if I butcher the MechE terms. Imagining a load-bearing metal beam. Every infinitesimal "slice" of the beam is going to have a specific vector pointing in some direction. However, you can slice it along different axes and each one yields a different force vector for the same point.
The tensor is then a mathematical object where each element is itself a multidimensional vector, so like, the tensor for the point (1,2,3) would contain 3 vectors for the shearing forces on the beam

does that sound about right?

>> No.12613715

>>12613639
Do you mean the series representation for [math]e[/math] or some general exponential function?

>> No.12613762

>>12613715
[math]\exp(z)=\sum_{k=0}^{\infty}\frac{z^{k}}{k!}[/math]

>> No.12613851

>>12613762
You can just use the ratio test to show it converges.

>> No.12614015

Does [math] x \mapsto f(x) [/math] mean the same thing in mathematics as [math] \lambda x. f(x) [/math] means in computer science?

>> No.12614037 [DELETED] 

>>12613851
>>12613639
Whithout proving the ratio test too, i fail to see how can that be a proof

>> No.12614063 [DELETED] 

>>12613851
>>12613639 (You)
Whithout proving the ratio test theorem too, i fail to see how can that be a proof

>> No.12614072

>>12613851
>>12613639
Without proving the ratio test theorem too, i fail to see how can that be a proof

>> No.12614162

>>12614072
Prove the ratio test theorem then

>> No.12614235
File: 125 KB, 1122x754, DIET 20.jpg [View same] [iqdb] [saucenao] [google]
12614235

Is this diet good for your brain?

>> No.12614438

This might sound retarded, but how do you learn theory?
Currently I'm learning math. I have no problem learning how to calculate something (linear inequality equation, system of two equations and similar).
I do have problem with memorizing words/sentences and want to improve my technique of learning. Currently I read something, close the textbook and try to rewrite what I just read on a piece of paper.

>> No.12614497

>>12597883
NO, NOT FUCKING REMILIAFAG
MY WIFE AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

>> No.12614554

how do I stop getting filtered by sign convention

>> No.12614656

>>12614162
I can't prove it, that why asking the
>qtddtot
here

>> No.12614735

>>12614072
>>12614656
So let me get this straight. You don't want to use a convergence test to test if a series converges?

Someone's certainly putting the stupid into /sqt/

>> No.12615133

>>12607198
>>12610530
https://en.wikipedia.org/wiki/Dictum_de_omni_et_nullo#Dictum_de_omni

>> No.12615747

can someone explain why -2^2 = -4 if exponentiation is defined like this:
[math]b^n = \underbrace{b \times \dots \times b}_{n\, \textrm{times}}.[/math]
-2*-2 is obviously 4

>> No.12615771

>>12615747
-2^2 != (-2)^2

>> No.12615836

>>12614735
How would you use a bleep-blorp to decum-etangle the manger-wobble, without knowing first what the bleep-blorp means?

>> No.12615949
File: 43 KB, 1014x552, ssdsegawf.png [View same] [iqdb] [saucenao] [google]
12615949

I don't get this, how do i know that angle D is congruent to angle C? it's not even corresponding or alternate interior.

>> No.12616234

>>12615836
If someone asks you if a series is convergent you kind of expect them to know what convergent means.

>> No.12616243

>>12614438
Start by learning to solve problems in general. Polya's book How to Solve it is a good one. Then get an easy proof based mathematics book and read the proofs and try and do some on your own. Or get a book on how to do proofs if you want more guidance.

>> No.12616272

>>12615747
> is obviously 4
Correct. Where did you get the idea it's -4??

>> No.12616296

>>12615747
()
^ right to left
implicit *, ~
*/& left to right
+-| left to right
-

>> No.12617468
File: 93 KB, 936x754, Screenshot 2021-01-23 100238.jpg [View same] [iqdb] [saucenao] [google]
12617468

M is the carrying capacity and N_0 the initial size. I don't understand what (<M/2) means after the N_0 here. Is it just saying the initial size is less than half the carrying capacity? I don't get why that has to be mentioned.

>> No.12617695

[Biology/Medicine]

I've always had this thing where I feel like my muscles never relax.
When I sleep it's like a pause button, not a reset button.

When I went to draw blood one day, the nurse said to relax my arm, but when I tried she said I wasn't relaxing.
> bruh.jpg
So I had to "force" my arm down to keep it straight so she could draw blood, she mentioned I had "something something" and I better tell my doctor about it, but I forgot what It was because I was trying to keep myself from passing out.

What can It be? The only times I've ever felt relaxed was during my two surgeries when I was under anaesthesia and when I tried some sleeping medicine from my ex, which was Stilnoct/Ambien.

>> No.12617698

>>12615949
nvm this, figured it out

>> No.12617745

>>12617468
>Is it just saying the initial size is less than half the carrying capacity?
Yes.

>> No.12617755

>>12617745
but why is that relevant, wouldn't that always be the case

>> No.12617761

>>12617755
Probably because in both cases it cannot be initially greater than M/2.

>> No.12617817

>>12616243
Thank you for your reply.
Maybe I wasn't specific enough in my question. I'm learning for a vocational part of exam that will be in more than a year from now (not sure how much theory I will be asked). I think I have a problem with learning technique.
I'm currently learning basic concepts of set theory and I believe I have problem with memorizing words like conjunction, disjunction and their respective operators.
I might also have problems with attention span when learning theory, I have no problems solving equations and such.

>> No.12618040

>>12613714
Yes, when you "slice" the metal beam in a certain direction, giving two components A and B, you're effectively asking yourself the question: what is the amount of force that B is acting on this face of A? Suppose the beam stands upright, and we make a thin slice which we again call A, then the stress that A experiences is the combined gravity -G of everything above it, divided by its area. It is purely perpendicular, so we now know that sigma_yy = -G, sigma_yx = sigma_yz = 0. If we were to make a vertical slice with the normal pointing in the x direction, we readily observe there is no normal stress (because there is no force that points sideways), and as for the shear stress, there is none either, because the slice attached to this surface is not dangling freely, but is firmly planted on the ground, so it is also zero. (In other words, if the two verticales slices weren't glued together, nothing would change. No force was necessary.) If we instead supposed we had a cube with one of its sides dangling slightly over the edge of the cliff, the only reason that mass is still hanging on is because the material it is clinging to the mass next to it. In doing so, it is exerting a downward shear force on that face.

I hope this explanation makes sense and gives you some intuition. If it's unclear I can make an MS paint drawing to show you what I mean.

>> No.12618046

>>12617468
It's a condition that makes the integrals make sense, otherwise the indices would be either misleading or wrong

>> No.12618059

>>12617761
>>12618046
thanks, i think i'm just overthinking it

>> No.12618077
File: 73 KB, 783x841, Untitled.png [View same] [iqdb] [saucenao] [google]
12618077

>>12618059
This should be a clearer treatment. Should be all right

>> No.12618082

>>12618077
wow thanks