Pic rel - /sci/ (#16691965) [Archived: 1202 hours ago]

Anonymous
6/8/2025, 7:16:43 AM No.16691965
Screenshot 2025-06-08 1.16.14 AM
Screenshot 2025-06-08 1.16.14 AM
md5: 4f16305cdb83b51f825c53d938203cc0🔍
The biggest problem I have with proofs is when the exercises involve more concrete problems. You see, it's like I can understand the theory behind something, but many times when it actually comes to applying that to concrete examples I have a harder time. Does that make sense? I'll come across a proof based problem, and if it's general enough, I can logic my way through the definitions and provide a pretty clear proof. But on the more concrete based examples I often don't even know where to start. It's really frustrating. How do I get better at this? Is this normal? Does anyone relate?
Replies: >>16692010 >>16692621 >>16694387 >>16694394 >>16694411 >>16695163
Anonymous
6/8/2025, 8:46:43 AM No.16692010
>>16691965 (OP)
Literally just show the definition in the set.
Anonymous
6/8/2025, 11:52:39 AM No.16692085
We want to show that f is in O(g), so show that the predicate on the right of the definition holds for f in particular.

We want to show that f is not in O(g), so show that the predicate on the right of the definition does not hold for f in particular.
Anonymous
6/8/2025, 9:42:06 PM No.16692621
60849952_p1
60849952_p1
md5: a673d2d1dc7aba45d762afab8b26c794🔍
>>16691965 (OP)
Before doing the problem you first need to understand the Intuition, for example what is [math]O(g)[/math]? I know it's defined right there but what is that definition really? Intuitively [math]O(g)[/math] is the set of all functions in [math]\mathcal{F}[/math] that grow as fast or slower than [math]g[/math]

What this problem is really asking you is to prove that [math]g[/math] grows faster than [math]f[/math]
At what point does [math]g[/math] outgrow [math]f[/math]? if you plug in some values you will realize this point occurs when [math]x>7[/math] (or [math]x \geq 8[/math] since [math]x[/math] is an integer)
So now we have the intuition we put it into a formal proof:
Assume [math]x \geq 8[/math]
[math]\Rightarrow x^2 \geq 8x \\ \Rightarrow x^2 \geq 7x+x \geq 7x+3 \\ \therefore g(x) \geq f(x) [/math]
This is true for all [math]x>7[/math]. So now set [math]a=7[/math] and [math]c=1[/math] and the condition inside [math]O(g)[/math] is satisfied (absolute value bars can be ignored here since [math]f[/math] and [math]g[/math] are always positive) so [math]f \in O(g)[/math]. This could be alternatively proved by letting [math]c\geq10[/math] then [math]f(x)\leq cg(x)[/math] for all [math]x \in \mathbb{Z}^+[/math].

To prove [math]g \notin O(f)[/math] assume the opposite for a proof by contradiction, so there exists some [math]a \in \mathbb{Z}^+[/math] and some [math]c \in \mathbb{R}^+[/math] such that [math]g(x) \leq cf(x)[/math] for all [math]x>a[/math]
So [math]x^2 \leq c(7x+3)[/math]
[math]\Rightarrow x^2 \leq 7cx+3c < 7cx+4c \leq 7cx+4cx = 11cx \\ \Rightarrow x^2 < 11cx \\ \Rightarrow x < 11c[/math]
And this should be true for all [math]x>a[/math], but it's also obviously false for all [math]x\geq 11c[/math], a contradiction. Therefore [math]g \notin O(f)[/math]
Replies: >>16692640 >>16694508
Anonymous
6/8/2025, 10:29:25 PM No.16692640
>>16692621
Ah thank you. This makes a lot more sense now. I guess I just don't have enough practice with these types of problems. The text book I'm reading has only had a handful of these types of problems and they were mostly at the start of the book so any intuition I might've built up about them a while back is just gone now.
Anonymous
6/10/2025, 6:50:09 PM No.16694387
>>16691965 (OP)
It's normal, and practice is key. The concrete follows from the abstract and is only a specific case, be sure to know plenty of specific examples that make things work.
Anonymous
6/10/2025, 7:00:50 PM No.16694394
>>16691965 (OP)
>How do I get better at this?
Practice.

>Is this normal?
Yes. Proofs, whether in math or physics, are a big step up in difficulty compared to the kinds of plug-and-chug problems you get used to in lower level classes. You transition from just seeing whether you can power through an equation or method to seeing whether you can actually bring multiple ideas together and make intuitive leaps in order to work through something where the process isn't already mapped out for you.

You'll get there, anon. Keep working at it.
Anonymous
6/10/2025, 7:22:15 PM No.16694411
>>16691965 (OP)
what the fuck is wrong with the person that wrote that definition of O(g)? just write english, the fuck is that horrible set theory notation
Replies: >>16694488
Anonymous
6/10/2025, 9:19:12 PM No.16694488
>>16694411
there are multiple instances of world's leading experts (fields medalists and others) who have written letters to the editor in famous math journals saying their previously published topic, on which they're an expert, was completely incomprehensible to them due to the overloaded (often idiosyncratic) notation. they mentioned that if they - leading experts - couldn't make heads or tails of the story, then who the fuck approved it, and that those authors need a bullet to the head (okay, i may be paraphrasing).
Anonymous
6/10/2025, 10:06:15 PM No.16694508
1737612094309071_thumb.jpg
1737612094309071_thumb.jpg
md5: 54d57c19c8c848aec1cf0aa5b1a15a78🔍
>>16692621
Good explanation. Thank you anon for contributing to society
Replies: >>16694545
Anonymous
6/10/2025, 10:36:38 PM No.16694545
>>16694508
This conveniently cuts the moment before a passing car backfires and the chimp rips the dog's head off.
Anonymous
6/11/2025, 12:03:41 AM No.16694656
OP here. Came across a similar (though much more simple problem).
>Let F={f|f:R->R}. Let S={(f,g) in FxF|There exists an a in R for every x>a(f(x)=g(x))}. Let f:R->R and g:R->R be the functions defined by the formulas f(x)=abs(x) and g(x)=x. Show (f,g) in S.
I wrote:
>Let a=0. Then for any x>a, f(x)=abs(x)=x=g(x). Hence (f,g) is in S.
Is this pretty much correct? One of the biggest problems I have is finding what to choose for an existential. This was an easy case but sometimes I don't know what to choose to make the statement work out.
Anonymous
6/11/2025, 4:10:44 PM No.16695163
>>16691965 (OP)
Logicking it out is usually just symbol pushing. There are limited things you can do in a lot of abstract proofs. For these kind of questions, you have to try and grasp the intuition behind the definitions. try and think about the notion this definition is formalizing.