Zulip Chat Archive

Stream: new members

Topic: Summation over 'some' subset of natural numbers


Anika-Roy (Jan 24 2025 at 19:24):

image.png
Hello!
I'm trying to formalize the theorem statement here but I'm running into issues in defining that there exists some not-predefined subset of the natural number for the summation. I'm attaching my code below, which possibly has many errors. Any guidance or criticims will be great!

import Mathlib.Topology.Algebra.InfiniteSum.Real
open BigOperators

/--
**Proposition**
Let `1 ≤ x < ∑' n, 1 / (n^2 : ℝ)`.
Then there exists a set `A ⊆ ℕ` such that `∑' n in A, 1 / (n^2 : ℝ) = x`.
-/
theorem subseriesDenseInBasel
    (x : )
    (h1 : 1  x)
    (h2 : x < ∑' n, (1 : ) / (n : )^2) :
     A : Set , ∑' n in A, (1 : ) / (n : )^2 = x := by
  sorry

Ruben Van de Velde (Jan 24 2025 at 21:03):

∑' n : A seems to parse, at least

Ruben Van de Velde (Jan 24 2025 at 21:04):

AA may legitimately be an infinite set here, right?

Jireh Loreaux (Jan 24 2025 at 21:25):

Indeed, for all but countably many values of x, A must be an infinite set.

Aaron Liu (Jan 24 2025 at 22:49):

It looks like docs#tsum doesn't support summing over Props, so ∑' n ∈ A, ... fails.

Edward van de Meent (Jan 25 2025 at 13:40):

you could write ∑' n, ∑ᶠ (_ : n ∈ A), (1 : ℝ) / (n : ℝ)^2 to skip the usual computability assumptions

Edward van de Meent (Jan 25 2025 at 13:40):

i believe the api is less than awesome tho


Last updated: May 02 2025 at 03:31 UTC