Zulip Chat Archive
Stream: ItaLean 2025
Topic: Projects: Sphere Packing
Pietro Monticone (Dec 09 2025 at 16:34):
This thread is dedicated to the Sphere Packing project:
- Repository: https://github.com/thefundamentaltheor3m/Sphere-Packing-Lean
- Informal & Formal Manager: @Sidharth Hariharan
Sidharth Hariharan (Dec 09 2025 at 17:30):
Useful links:
- Repository
- My thesis
First objective: Proposition 4.4.6 using preceding lemmas in Section 4.4.1. We will state (but sorry out) Lemma 4.2.3, a direct dependency of the lemmas preceding Proposition 4.4.6.
Sidharth Hariharan (Dec 09 2025 at 17:30):
Draft PR: spherepacking#229
Tito Sacchi (Dec 09 2025 at 23:43):
I fixed some issues with the definitions we wrote today and started writing a proof of the first integral identity. I pushed the code to my fork, any advice is welcome! We can look at the code together tomorrow.
Sidharth Hariharan (Dec 10 2025 at 15:36):
General resources
- Project website
- Blueprint
- GitHub Repository
- GitHub Projects Dashboard
- Project Zulip channel (on the Leanprover Zulip)
- Viazovska's talk at Big Proof (where our repository was opened up for public contributions)
- My Master's thesis (a version that outlines our strategy for formalising the construction of the magic function)
Tito Sacchi (Dec 11 2025 at 11:48):
image.png
Moving forward with Cauchy-Goursat :+1:
It is quite painful to prove identities with I_2 and I_4 because your definitions use parametrizations such as z_2', z_4', but our formalization of Cauchy-Goursat uses interval integrals directly. We basically need another change of variables.
Michael Rothgang (Dec 11 2025 at 13:22):
We talked about this in person: feel free to share if what we discussed (providing a specialised version of change of variables as an auxiliary lemma, for now) helps.
Tito Sacchi (Dec 11 2025 at 13:27):
Thanks! Currently working on rewriting the integrals using some identities available elsewhere. I'll prove a general lemma for simple linear changes of variables next.
Tito Sacchi (Dec 11 2025 at 13:59):
lemma const_add_variable_change (x₁ x₂ x₁' : ℝ) :
∫ t in x₁..x₂, f t =
∫ t in x₁'..(x₁' + (x₂ - x₁)), f (t - (x₁' - x₁)) := sorry
This should do the job and should be provable fairly easily using the standard Mathlib API.
Last updated: Dec 20 2025 at 21:32 UTC