« first day (1197 days earlier)      last day (1286 days later) » 

00:15
@shintuku Yes, that's exactly right. "∀x∈ℕ(¬x<x)" means "∀x∈ℕ(¬(x<x))".
@F.Zer thanks!
You're welcome !
 
2 hours later…
01:50
Does ∀n ∈ N(n = 0 <-> -n = 0) need to be an axiom? Otherwise, we can't guarantee that -n ∈ N, so we can't use Peano's axioms
We could also maybe use ∀n ∈ N(n = 0 <-> -n ∈ N) as an axiom
We would of course posit the definition -n := n + -n = 0
oh, I found a work-around: ∀n ∈ N(-n ∉ N <-> -n ≠ 0)
hm... the best I can do is ∀n ∈ N(-n ≠ 0 -> -n ∉ N) via contradiction, but the biconditional is out of reach
also my proof header is getting HUGE lol
It actually began looking like an actual python file... which makes me wonder: is there a programming language in which we can easily implement your system? I've heard there are proof systems out there
02:13
@shintuku Negation "–" is not a symbol defined in PA.
yeah, I tried getting around that with -n := n + -n = 0, but I'm not getting very far
@shintuku If you work within PA, you can't use that symbol. Look at the axiom ∀ x ∈ ℕ ( x < y ⇒ ∃z ∈ ℕ ( x+z = y ) ).
hm... dang... how would you incorporate negatives?
@shintuku Why would you want to incorporate negatives ?
I need the statement ∀n ∈ N ¬∃k ∈ N (n≠0 -> n = -k) for my proof of bijection between the integers and natural numbers
02:23
@shintuku Oh, I don't know about that proof. FWIW, @user21820 added the ℤ axiom ∀ x ∈ ℤ ( x ∈ ℕ ∨ –x ∈ ℕ ). Not sure if that is related to your goal.
oh cool! this is exactly what I need
do you know if he has other Z axioms? (and Q or R axioms?)
in any case thank you very much for ∀ x ∈ ℤ ( x ∈ ℕ ∨ –x ∈ ℕ ), this solves my problem
@shintuku We discussed ℤ axioms, here:
Jul 23 at 11:01, by user21820
For integers, take all the axioms of PA− (from my post) except the last one, and change "ℕ" to "ℤ", and you will get axioms for ℤ. Add the binary operation − : ℤ^2→ℤ, and the axiom ∀x,y∈ℤ ( (x−y)+y = x ). Add the axioms ∀x∈ℕ ( x∈ℤ ) and ∀x∈ℤ ( x∈ℕ ∨ −x∈ℕ ). And you're good to go!
omg, thank you so much! this is exactly what I was missing
Jul 23 at 12:40, by user21820
Now ℤ is supposed to extend ℕ; it includes every member of ℕ plus 'negative' ones. Of course if we want to have stuff less than 0 the last axiom cannot hold if "ℕ" is changed to "ℤ". It turns out that we can keep all the axioms of PA− for ℤ simply changing "ℕ" to ℤ" except the last axiom.
@shintuku You're welcome !
maybe we should make a google document with these, no? these are super useful
02:30
Yes. For practical reasons, I copy each axiom in a text file.
I should go, now. See you next time !
see you! thanks!
02:48
@F.Zer Rare occasion, but wikipedia can help you.
You do need a fact that you haven't proven yet, namely ∀x∈ℝ ( x ≥ 0 ⇒ ∃!y∈ℝ ( y ≥ 0 ∧ y·y = x ) ), which is needed to define sqrt(x) for each non-negative x. But for now assume it without proof, and you are free to use sqrt on non-negative reals.
@shintuku As F.Zer said, that's correct. The standard precedence rules are (highest to lowest): [infix operations],[relations including =],¬,∧,∨,{⇒,⇔}.
So "¬ x+y < z" would mean "¬( (x+y) < z )".
Both ⇒,⇔ are usually not distinguished in precedence, so I grouped them together.
@shintuku There are proof systems out there, but the established ones (e.g. Mizar,Coq) aren't Fitch-style. It should be a relatively simple programming project to implement my system, but I never got around to doing it.
 
4 hours later…
06:48
Alright
I think I got it
Prove that:
F = {⟨x,y⟩: x ∈ N ^ y ∈ Z ^ φ(x,y)}
φ(x,y) <-> ∃z∈N [(x=2z ^ y=z) v (x=2z+1 ^ y=-z)
define a function
07:42
alright: this was considerably harder than I thought, but here it is:
Prove
F = {⟨x,y⟩: x ∈ N ^ y ∈ Z ^ φ(x,y)}
φ(x,y) <-> ∃z∈N [(x=2z ^ y=z) v (x=2z+1 ^ y=-z)
defines a function
\\Header (and we assume the rules in the MathSE post)

PA1 ∀k∈N ∃m∈N ( k = m·2 ∨ k = m·2+1 )
PA1* ∀k,m ∈ N (k = 2m -> ¬∃n ∈ N (k = 2n+1)) \\modified PA1
P1 Z ∈ set
P2 φ is a property with two parameters and ∀x∀y: φ(x,y) <-> ∃z∈N [(x=2z ^ y=z) v (x=2z+1 ^ y=-z)]
P3 y ∈ {x : ψ(x)} <-> ψ(y)
P4 N ⊆ Z

\\Prove F := {⟨x,y⟩:x∈N ^ y ∈ Z ^ φ(x,y)} is a set in P(N x Z)
1 N x Z ∈ set \\from P1, **naturals**, **product-type**
2 F = {p: p ∈ N x Z ^ ∃n ∈ N ∃z ∈ Z (p = ⟨n,z⟩ ^ φ(n,z))} ∈ set \\from **comprehension**
also, I'll look into Mizar and Coq, thanks for the tip!
you were very right that I had shortened the proof too much, I actually had not fully understood the mechanics of the proof
07:56
Is there any way to shorten it? maybe some theorems that would be of interest?
 
4 hours later…
11:56
@shintuku FWIW, look at ∃elim rule; in line 7 you write "m := n = 2m ∨ n = 2m+1". However, using this system, should be "Let m ∈ ℕ such that n = 2·m ∨ n = 2·m+1". In this system, you should declare the type of every variable.
@user21820 Thank you. I'll look into it.
 
2 hours later…
14:04
∀ a,b,c,x ∈ ℝ ( a·x^2+b·x+c = 0 ∧ a ≠ 0 ⇒ x = –(b/(2·a))+√((b^2-4·a·c)/4·a^2) ∨ x = –(b/(2·a))–√((b^2-4·a·c)/4·a^2) ) [lemma]
  Given a,b,c,x ∈ ℝ:
    If a·x^2+b·x+c = 0 ∧ a ≠ 0:
      a·x^2+b·x+c = a·(x^2+b·x/a+c/a)
      a·(x^2+b·x+c) = a·(x+(b/(2·a)))^2+c/a–(b^2/4·a^2))
      x^2+b·x+c = (x+(b/(2·a)))^2+c/a–(b^2/4·a^2) [cancellation since a ≠ 0]
      (x+(b/(2·a)))^2+c/a–(b^2/4·a^2) = 0
      (x+(b/(2·a)))^2 = –c/a+(b^2/4·a^2)
      (x+(b/(2·a)))^2 = (b^2-4·a·c)/4·a^2
      √((x+(b/(2·a)))^2) = √((b^2-4·a·c)/4·a^2)
@user21820 I proved quadratic formula. Used the fact "a ≠ 0" to perform the cancellation. Could you tell me what do you think ?
 
1 hour later…
15:09
If (y-ax)(y-bx)(y-cx)=0 then Prove that d(y)/dx = y/x.
 
7 hours later…
21:43
@F.Zer oh you're right, thanks for the tip!

« first day (1197 days earlier)      last day (1286 days later) »