[ 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: 82 KB, 800x450, eca.jpg [View same] [iqdb] [saucenao] [google]
9809586 No.9809586 [Reply] [Original]

>Oldschool compsci papers:
>We did this shit because we have good reasons based on this and that and here is the mathematical proof.

>Newschool compsci papers
>uhmm liek... we used machine learning i dunno why lol but it just kinda werks lol (it doesnt work)

>> No.9809605

Proofs? Who are you quoting?

>> No.9809623
File: 156 KB, 887x1128, Curriculum 68 Recommendations for academic programs in computer science a report of the ACM curriculum committee on computer science.png [View same] [iqdb] [saucenao] [google]
9809623

>>9809586
>Oldschool compsci curriculum
>pic

>Newschool compsci curriculum
>python, java, OOP, data structures 1&2, algorithms, SQL, system programming in C, lisp, networking, web design, software engineering, discrete math 1&2, mobile app dev, meme learning

>> No.9809813

>>9809623
While it doesn't have the analysis requirement (though, I double major in math and CS so I'm doing it anyway), my school's CS curriculum is nearly identical to this..

>> No.9809910

>>9809813
Same. We do have the databases and networking classes though.

>> No.9809937
File: 332 KB, 440x330, LAUGHING.gif [View same] [iqdb] [saucenao] [google]
9809937

>>9809910
Same here, but I think we have some of those classes to appeal to those who want to minor in CS, but all the classes that count for the major are tied in identical prerequisites. Even numerical analysis requires multivariable calculus and heavily recommends taking an ODE class

>tfw your CS curriculum is both employable AND not a meme

>> No.9809988

>>9809937
which uni?

>> No.9810195
File: 32 KB, 484x426, 80B5C42A-88D3-4EDE-81C4-8950C38610FE.jpg [View same] [iqdb] [saucenao] [google]
9810195

>>9809988
A certain public research university near but not within New York. Respectable but not top 10. I hope that with good grades, a double major, good research, and some job experience, I can do research at a good place ;-;

>> No.9810329

>>9809623
Now, I know that most of the his is a meme but
>complaining about a dedicated advanced algorithms class
Why are you complaining about entropic extractors asymptotic tricks. People here complain a lot about the lack of mathematical rigor in CS, and now you’re complaining about an area of undergrad education that actually employs a good amount of it.

>> No.9810458

>>9810329
*entropic extractors and asymptotic tricks

>> No.9810465
File: 35 KB, 196x361, weird_goblin.png [View same] [iqdb] [saucenao] [google]
9810465

Oldschool CS papers:
>This will definitely get us laid this time, we've been virgins for way too long
Newschool CS papers:
>Here are the best brands of thighhigh and stockings on the market. Also don't forget to clean out your chastity cages! :3333
What went wrong?

>> No.9810481

>>9810465
>This will definitely get us laid this time, we've been virgins for way too long
This is true for every STEM major

>> No.9810521

>>9809910
They had databases too:
>Information Organization and Retrieval

>> No.9810526

>>9810481
If you can't fuck Stacy become Stacy and Chad will rail you.

>> No.9810616
File: 40 KB, 600x456, B7380F24-5B34-48BC-93ED-AE15BA94950F.jpg [View same] [iqdb] [saucenao] [google]
9810616

>>9810465
A lot of my old school CS professors are chads who are also very knowledgeable on the EE side even though they hold degrees on the CS side. I have yet to meet an OS/systems and architecture professor who isn’t into RF and amateur radio

>> No.9810621

>>9810465
OwO what's this

>> No.9810705

>>9810465
Internet

>> No.9810757
File: 90 KB, 566x683, A747C526-88E7-4FE5-B9FC-544B4130265F.jpg [View same] [iqdb] [saucenao] [google]
9810757

>> No.9810787

>>9810465
Poor CS lads actually never got laid

>> No.9810917
File: 53 KB, 280x403, 3AE47C95-4093-4C2E-A8DA-0BBF8193DC5B.jpg [View same] [iqdb] [saucenao] [google]
9810917

>>9810787
Tbh undergrad CS has the the a ratio of incompetent, messy slobs to clean, social yet equally as incompetent chads closest 1 out of any other major I’ve seen. You have the cool people who are competent and lie somewhere in the middle, but they are gems in a sea of students who made CS into a meme

>> No.9810947

>>9809623
Looks like the curriculum for bachelor + first year of master here in datalogi (compsci for non-brainlets), kinda weird that the pic doesnt include mathematical logic
t. Sweden

>> No.9811444
File: 85 KB, 1024x819, A12462ED-0A64-44C0-97F2-7AA0D6CC8095.jpg [View same] [iqdb] [saucenao] [google]
9811444

Analytical combinatorics is slowling becoming hot in the field of algorithms. Hopefully, in the future, they will require at least one analysis class from undergrad.

>> No.9812450

>>9809586
People on this board hate CS so much.
It gets old.

>> No.9812458

>>9809586
How are you interpreting no longer needing to explicitly define how a program solves a task as anything other than progress?

>> No.9812591

>>9812450
CS and CS students are garbage most of the time.
It gets old.

>> No.9813046

>>9812591
I can see CS students being garbage, but CS is a fine field of study. Not gonna lie, it’s pretty impressive that a field/specialization could blow up in only 60 years like CS did.

>> No.9813051

>>9812591
That's just salt though.

>> No.9813268

>>9812450
I'm a CS grad student and even I sometimes hate CS kids. Most of the fuckers that are in the department gives no shit about automata theory, math, or runtime complexities and optimization. They're the typical lol just use JavaScript or use python dudebros who got in because they think they can get 100k starting. They also don't take any advanced and in depth CS courses such as digital logic, compilers, OS etc and cruise by with a degree with shitty web dev and Java courses. It's a fucking disgrace.

>> No.9813315

>>9813046
You mean, the field that mathematicians developed and advanced, then got turned into a "major" for people too dumb for mathematics to go into so they can fix bugs in shitty bank software

>> No.9813322

>>9809623
The program in your pic is literally the program of most european universities.

>> No.9813335

>>9813268
the funny thing is that the people who get into CS only for the mythical 100k starting salary are the ones who are least likely to get it.

>> No.9813513

>>9813315
You have a chip on your shoulder.
I see hundreds of CS students who don’t care, but the ones who actually care go out of their way to take hard classes in both the math and CS departments. Your shitposting makes you seem more bitter than your average CS student is stupid.

>> No.9813518

>>9813268
I agree that I saw a lot of these people in my early undergrad. Once I started taking the classes you have to go out of your way to take (graduate algorithms, OS, algebraic combinatorics, etc.) I found passionate CS students. Script kiddies seem sort of an arbitrary thing to mock in hindsight.

>> No.9813529

>>9810917
>>9813268
>>9813315
CS undergrads are scum, the only ones who are somewhat intelligent and able to do research are the ones who should be in maths but decided against it for some stupid reasons.

I mean, Maths undergrad programs have more complete freaks, but the average student is less incompetent than a CS one

>> No.9813534

>Future compsci paper
>I smooshed together all these neural nets because I was bored and now it's sentient. Please help, I don't know how this happened.

>> No.9813577

>>9813529
Or you know, they take mathematics and computer science. Double major or dual degree.

I see smart double majors who usually take a physical science and math, a physical science and CS, or CS and math. I don’t understand mathfags and their insistence that their single major is the best when you can pick TWO things you like and study them ;^)

>> No.9813580

>>9813534
From my experience, it’s more
>computational circuits
>kolmogorov complexity intensifies

>> No.9813603

>>9813577
Not all universities have double majors. And dual degree means you need ~5 years for your bachelor degree alone, why would anybody want that

>> No.9813609

>>9813603
I didn't do a dual degree, but I did do a double major. It's the perfect crime. While you won't hear it on this board, math majors have written some of the most disgusting code I've ever written, and that's not even the CS part of CS. They also fail to translate the merits of their formal education into good training. I see a lot of them talk about how they could do software jobs, but none of them actually write something that works in their spare time. Hell, they don't even any scientific code, and the times their classes have asked them to do so, it's been met with bitching (doing rudimentary MATLAB assignments in vector calculus and ODE classes comes to mind)

As for the sins of CS students, well, this is what /sci/ bitches about. They're not exempt from hate, but I can stand stupidity much more easily than the self-righteousness of your average /sci/ math major who thinks he's hot shit.

>> No.9813613

>>9813609
*have written some of the most disgusting code I've ever read

*don't even write any scientific code

Urgh, math-only majors got me into a frenzy

>> No.9813621

Modern CompE & EE grads are much closer to what computer science was in the 80s and 90s than it is now.

>> No.9813624

>>9813621
So they are obsolete?

>> No.9813628

>>9813624
No, they come out with more breadth and depth on computer systems and how things work at a more concrete level instead of doing

import DoMyWorkForMe
DoMyWorkForMe.run()

>> No.9813641

I used to think engineers got it hard, but after auditing a few classes, swapping homework during break, etc., that's actually not the case. Your modern CompE and EE is usually just a Cadence + MATLAB monkey.

>>9813621
I've seen EE and CompE people in my CS classes. Funnily enough, they actually score much worse across all the classes I've had with them. So thanks for the curve ;^)

>>9813628
Not true dude. In my computer architecture class, we had an assignment to translate an equation into assembly that detected overflow. The 4 EE people in my class just used GCC to generate the code even though it was explicitly forbidden. They ended up getting caught and getting a 0 for the assignment. For the rest of the year, they kept copying assignments off of each other. 2 of them got removed from the class after they got caught a second time.

As for the depth of computer systems...3 days before the test I was teaching the material to my peers as my own review. None of them had a damn clue what was going on.

>> No.9813645

>>9813641
Meh, maybe it's different school to school, then. CompE is a very variable program. My school of literally 40,000 undergraduates only had 22 CompEs finish because it was designed to be rigorous and root out morons and cheaters.

>> No.9813647

>>9813641
Also...
>we had an assignment to translate an equation into assembly that detected overflow
Really? Some EEs couldn't do that? Even the most banal RISC architecture has overflow detection in the status register. It's two or three lines of assembly.

>> No.9813648

math i win

>> No.9813669
File: 347 KB, 492x500, 1528239482524.png [View same] [iqdb] [saucenao] [google]
9813669

>>9813647
Yeah, it was a disappointment to say the least. CS is non-memey my school, and while we typically don't get to hardware design, we get a good amount of switching theory and digital logic design in our architecture classes. Our big topics were:

Numerical representation (IEEE floats, 2's comp numbers, bit shift tricks, some hardware algorithms like Booth's multiplication)

Assembly (compilation basics, hardware specific instructions, translating both ways, stack building, etc.). We used a subset of x86 rather than RISC since we used linux machines to write our stuff

Digital logic design (your typical transistors, gates, combinations, K maps, tables, rewrite rules, latches, clocks, etc.)

Caching (L caches, methods of caching, writing a cache, measuring efficiencies, etc.)

I think that, for a course that was supposed to be an intro to the low abstraction courses in CS, it was a good curriculum with challenging projects

>>9813645
Our CS has an easier route and the hard route. The easy route has some software engineering classes that people don't take seriously, etc. However, the hard route that requires a lot of math prerequisites are a lot of fun and take up a lot of time. The people who end up on this track typically break into either the big companies or into the prestigious universities for graduate education.

>> No.9813690
File: 30 KB, 514x536, 1442353169468.jpg [View same] [iqdb] [saucenao] [google]
9813690

>>9813529
>the ones who should be in maths but decided against it for some stupid reasons

I swear, /sci/ chooses their studies based on looks instead of passion, it's insane.

Why are you all so empty inside?

>> No.9813780

>>9813609
Thanks for the blog man.
>(*SNAP*)

>> No.9813787

>>9813690
A good amount of this board is full of undergrads. The underclassmen are excited that they chose the "one true major" that will lead them as either intellectual or industry leaders. The upperclassmen usually don't have exposure to other (or hell, even neighboring) fields, so they end up thinking they made the "right" decision. I've seen this from both engineering and traditional math/science students. Respect for the other side doesn't come until they work in the field with other people who are intelligent yet who

*gasp*

chose different major paths than "that major that means you're one of the smart ones™." Sometimes, if the field is especially young, like CS, they won't even respect them in the field even though any look at a simple graduate text in the field indicates that it is a field worth of study. The reckon that the lack of self awareness happens when you spend a lot of time in a room trying to understand complicated material. A lot of what this board shittalks seems to be a mix of legitimate frustration and pleas of academic validation.

>> No.9813789

>>9813780
No problem man, and make sure to smash that subscribe button for more content ;^)

>> No.9813792

>>9813335
Nah, that's definitely not true. People who go to top 20 cs programs with that mentality are probably making 200-300k in their late 20s. Google and the likes hire almost all of their employees from top 20 programs anyways.

>> No.9814029

>>9813787
stop. this isn't a redpill thread, delet this

>> No.9814033

>>9809623
>computer graphics
>1968

>> No.9814041
File: 5 KB, 344x146, images.jpg [View same] [iqdb] [saucenao] [google]
9814041

>>9809623
Damn, you got me to a fucking T.
I'm sorry I didn't study in high school.
If it dosen't pan out, I can just kill myself.
Or work at Kroger.

>> No.9814043

>brainlet
CS major
>patrician
Math or EE major with CS grad degree

>> No.9814058

>>9814043
>brainlet
/sci/ poster
>patrician
math with CS grad
EE who catches up on math and takes CS grad
CS who catches up on math and takes S grad
and best of all:
CS and math double major undergrad who does CS or math in grad

>> No.9814071

>>9809623
Learning CS right now with something that looks like your pic. Stay mad, mutt.

>> No.9814076

>>9814043
>>9814058
Im doing EE major with math diplomate.
Im a pleb tier or patrician?

>> No.9814079

>>9814058
>CS and math double major

These are terrible. CS+math combos always take the most brainlet math courses to complete the major.

>> No.9814083

>>9814079
maybe at your school

>> No.9814092

>>9814083
Or you're just a brainlet not realizing how trivial your studies are.

>> No.9814107
File: 51 KB, 400x323, 1527814403108.jpg [View same] [iqdb] [saucenao] [google]
9814107

>>9814092
This.
Why would you let yourself study math that is unrelated to your course to become a mathatician?

>> No.9814109

>>9814033
>hurr durr all graphics are real time

>> No.9814111

>>9814079
Imagine having the discredit those who do more by saying they take the easy classes

I got into graduate automata and coding theory because I took the abstract algebra classes. I took analysis, topology, stochastic processes, and I’m currently in PDE’s, advanced algorithms, compiler design, and I got special permission to do a quantum mechanics class (admittedly I’m gonna have to do some catching up). I don’t know what you’re on about. People can do more than one major, and quite frankly, if you’re not in an ABET engineering curriculum that forces you to take a million prerequisite classes your first two years, doing a double major and taking the hard classes means that you’re lazy.

>> No.9814113

What should I study if I want to work with computer vision and shit, EE, CS or Math?

>> No.9814115

>>9814113
EE with some CS machine learning classes are great. Alternatively, CS with architecture/low abstraction courses and machine learning are just as great. Pay close attention in multivariable calculus, probability theory, and linear algebra.

>> No.9814120

>>9814111
*not doing a double major and taking the hard classes means you’re more than likely lazy

>> No.9814125

>>9814107
>he doesn’t enjoy studying math for its own sake
>he doesn’t realize that all areas of math come in handy at one point or another, and having a good breadth of knowledge can only help if studied properly
>he doesn’t realize that despite all this, learning how to apply the math is just as rewarding

Imagine bragging about rejecting knowledge and calling *other* people the brainlets.

>> No.9814157
File: 47 KB, 522x472, %22he went from an s to an m ;^)%22.png [View same] [iqdb] [saucenao] [google]
9814157

>>9814079
This argument doesn't hold up under scrutiny, though. Most people do the Math + CS double major because they feel like they want an in depth understand of CS while getting a full mathematics education.

Suppose they took the most brainlet CS courses. Suppose their school didn't offer compilers, automata, and advanced algorithms. Suppose instead that they were offered webdev and software engineering classes. At my school (and at all others I've seen), you take the standard calculus sequence, into to linear algebra, and then an intro proof course. You then have to take analysis, probability theory, and 8 other upper level electives like topology, abstract algebra, differential geometry, etc. This board talks about the merits of a mathematics education, and seeing as how math education in undergrad isn't too different across different schools, your argument reeks of bitterness. You're not somehow superior to a person who double majored in your subject and something else because they took "muh weaker courses."

>> No.9814160

>>9814125
I didn't even meant to call you a "brainlet" I was just asking why.

>> No.9814173

>>9814160
My research interests lie in the intersection of theoretical physics and theoretical CS. That means I'm going to be undertaking a lot of algebra, a lot of analysis for PDE's and analytical combinatorics/algorithms, quantum mechanics, information theory (which has its roots in thermodynamics and statistical mechanics), etc. It's both useful and piques my interest.

>> No.9814227

>>9814076
Diplomate? Do you already hold a math degree?
The secret is that if you like what you're doing, can make a decent living off of it, and feel like you're using a good amount of your brain doing it, then you're more than likely doing the right thing. Don't let undergrads on /sci/ tell you otherwise

>> No.9814229

>>9814227
I didnt want to say diplomate because a diplomate is something different everywhere but its less bad than calling it a minor.
Its basically a "bonus" degree you get alongside your main degree when you graduate for taking a year and a half of pure math classes.
You make a good point, im passionate with what im studying right now, I just wanna know if im wasting my time with the diplomate.

>> No.9814238

>>9814229
Nah, if you enjoy it and feel like you're learning a lot, then you're capitalizing on your education.

>> No.9814335

>>9814157
>tfw differential geometry is grad level at my school

>> No.9814368

>>9814335
I said upper level, not grad level. By upper level, it means "both your proof class and real analysis class are required, which means you are expected to do some rigorous proofs on your exams."

>> No.9814370

>>9814335
Then take the graduate course desu

>> No.9814999

>>9814335
What school?

>> No.9815323

>>9814999
http://catalog.unc.edu/courses/math/

>>9814370
I'm planning to

>> No.9815336

>>9815323
Yeeesh, I'm looking at your math and CS curricula. Correct me if I'm reading it incorrectly, but are like..half of your math and CS classes that are considered upper level just graduate courses at your university? I see a 400s level web development class and that does not inspire confidence, since at my school, 400s indicates that it's usually a senior level class.

>> No.9815350

>>9815336
600+ is grad level
400 is so shitty Business school grad people can still take those undergrad classes

>> No.9815422

>>9815350
I'm so sorry. On the bright side, if you do really well in your courses, look here (and to other universities) for research, and get some good recommendations, you can get into really good schools after undergrad!

>> No.9815441

>>9815422
what are you talking about. The school's fine. Read the curriculum, It starts out with 401, and you need a bunch of 500+ classes to graduate. The numbers are inflated to accommodate bizfag. There's literally nothing stopping you from taking OS, or compilers as even a 2nd year.

>> No.9815447

>>9815441
My bad. Damn though, 400s until you get to the classes for the major. That's the first time I've seen that.

>> No.9815448

>>9809586
Just like C is better than "C++", math is better than CS, aka "math++".

Anyone who has taught or TAed a course on algorithms, automata theory, or mathematical logic can tell you that CS majors are in no way élite, and most can get only get 50% on automata theory exams involving basic reductions and use of the pumping lemma. This is true even at selective schools such as MIT and Harvard.

>> No.9815458

>>9815448
this is ironically accurate
t. automata TA

>> No.9815459

>>9815458
fuck, i meant unironically. I've been shitposting in too many threads. It's time to stop

>> No.9815461

>>9815448
In my experience, students like that tend to stay until data structures or computer architecture. In my automata class, the average was around 85% since the people in those types of classes wanted to be there. In fact, people did more poorly on basic probability quizzes than in my cryptography class when we covered randomness extractors for some keygen. I honestly think that good CS students get a lot of shit just because the name of what they chose to major in was "CS," even if they take enough math to double major or minor.

>> No.9815475

>>9815448
>Just like C is better than "C++",

>>>/g/tfo

>> No.9815491

>>9815448
>>9815458
While it's true that everyone and their mother wants a CS degree for the fabled employability, I still see a lot of people in both CS and math classes who are legitimately interested in doing mathematically charged research. They double major or take a lot of math classes nonetheless. The ones that care and do well are actually way more interesting people than a lot of people I've met in my math classes to be completely honest, and I double major in both.

>> No.9815726

https://www.cs.utexas.edu/users/EWD/ewd10xx/EWD1016.PDF

Researchers like Dijkstra still exist. Also
http://www.gigamonkeys.com/knuth-dijkstra/

>> No.9815849
File: 102 KB, 979x487, 1521492872950.jpg [View same] [iqdb] [saucenao] [google]
9815849

>>9815726
Nice website.

>> No.9815864

>>9810757
a fellow berkeley fag I see

>> No.9815868
File: 73 KB, 843x332, 0198.jpg [View same] [iqdb] [saucenao] [google]
9815868

>>9815849

>> No.9815902

>>9815868
Dijkstra's arguments are in favor of a rigorous Computer Science that still exists. Your average CS dork may not get it, but CS is a fine field.

>> No.9815935

>>9813787
desu I'm a grad student and I still think biology, geology and similar are fucking memes

>> No.9815940

>>9815448
As a student, I'll point out that loads of new subject matter gets introduced in a CS course every semester. The CS exams are on par with math, and both are pretty hard. Math gets a lot more reuse of the same stuff, with CS every course is building anew it seems. The theory parts are similar in difficulty.

You're a crap TA btw, good luck getting tenure.

>> No.9815965

>>9815940
As a student of both CS and math, I'd say that math edges CS as far as difficulty goes for the first 2 years, but once you get past basic algorithms and go into your advanced classes (read: the classes that themselves have hard math prerequisites) things get really tough. Those problems are tough and a lot of fun to solve.

>> No.9815966
File: 8 KB, 363x364, 1528364655454.png [View same] [iqdb] [saucenao] [google]
9815966

>>9815965
>>9815940

>> No.9815973

>>9815966
Fantastic post.

>> No.9815978

>>9815966
I don't know your angle, but you're trying too hard.

>>9815965
Math is a bit harder, but I wouldn't say either is too easy. And both are fun end-to-end.

>> No.9815981

>>9815966
>"real CS is tackled by a math student, and it's usually graduate level"
>CS student who takes a lot of math or double majors, takes the same graduate level classes
>"lol CS is a joke, anyone can do it"

And I'M the one baiting, apparently.

>> No.9815984

>>9815978
I dunno. I see both sides and gotta say that it's not too different. Making the distinction between math and computer science at an undergraduate level makes sense, but after starting analytical combinatorics and algorithms, I don't really think CS is easier than math. Then again, I take some of the graduate courses at my university, so maybe that's different than your average CS class

>> No.9816035

>>9815981
>goes to a shitty school for math
>hey math is just as easy as cs

brainlet plz.

>> No.9816039
File: 2.97 MB, 2200x3276, CS school.png [View same] [iqdb] [saucenao] [google]
9816039

>>9809586

>> No.9816044
File: 85 KB, 957x466, codemonkeys BTFO.jpg [View same] [iqdb] [saucenao] [google]
9816044

>>9815726
How can one man be so based?

>> No.9816083

>>9816035
CS is not uniformly trash across the nation. Also my school is within top 20 for math.

>> No.9816277
File: 137 KB, 434x500, image.jpg [View same] [iqdb] [saucenao] [google]
9816277

>>9809623
> tfw no core math subjects in my “compsci” major