proofs
Provedattentioncertified-onlineaffine-information-frontier

Grid indexing constructs the scalar whole-cell certificate

For all x,anchor,width∈R\htmlData{mx=1}{x, \mathit{anchor}, \mathit{width} \in \htmlData{mx=0}{\mathbb{R}}}, if 0<width0 \mathrel{\htmlData{mx=1}{<}} \htmlData{mx=0}{\mathit{width}}, then

(∣x−(anchor+width⋅(⌊x−anchorwidth⌋+12))∣≤width2)\left(\htmlData{mx=16}{\left\lvert \htmlData{mx=15}{\htmlData{mx=0}{x} - \left(\htmlData{mx=14}{\htmlData{mx=1}{\mathit{anchor}} + \htmlData{mx=13}{\htmlData{mx=2}{\mathit{width}} \cdot \left(\htmlData{mx=12}{\htmlData{mx=10}{\htmlData{mx=9}{\htmlData{mx=8}{\left\lfloor \htmlData{mx=7}{\frac{\htmlData{mx=5}{\htmlData{mx=3}{x} - \htmlData{mx=4}{\mathit{anchor}}}}{\htmlData{mx=6}{\mathit{width}}}} \right\rfloor}}} + \htmlData{mx=11}{\frac{1}{2}}}\right)}}\right)} \right\rvert} \mathrel{\htmlData{mx=19}{\leq}} \htmlData{mx=18}{\frac{\htmlData{mx=17}{\mathit{width}}}{2}}\right)
What the notation means · or point at any symbol
RReal numbers
The real numbers: every point on the number line. (ℝ)
x ∈ TVariables
x, anchor, width range over real numbers: the statement must hold for every choice. ((x anchor width : ℝ))
<Less than
0 is strictly smaller than width. (0 < width)
|x|Absolute value
The size of x − (anchor + width · (⌊(x − anchor)/width⌋ + 1/2)) without its sign. (|x-(anchor+width*(((⌊(x-anchor)/width⌋ …)
≤At most
|x − (anchor + width · (⌊(x − anchor)/width⌋ + 1/2))| is less than or equal to width/2. (|x-(anchor+width*(((⌊(x-anchor)/width⌋ …)
·Multiplication
width times ⌊(x − anchor)/width⌋ + 1/2. (width*(((⌊(x-anchor)/width⌋ : ℤ) : ℝ)+1…)
(x : T)As real numbers
⌊(x − anchor)/width⌋ read as an element of real numbers. Mathematicians leave this implicit; Lean needs it to know which kind of number is meant. (((⌊(x-anchor)/width⌋ : ℤ) : ℝ))
+Addition
⌊(x − anchor)/width⌋ plus 1/2. (((⌊(x-anchor)/width⌋ : ℤ) : ℝ)+1/2)
(x : T)As integers
⌊(x − anchor)/width⌋ read as an element of integers. Mathematicians leave this implicit; Lean needs it to know which kind of number is meant. ((⌊(x-anchor)/width⌋ : ℤ))
⌊x⌋Floor
The largest integer not above (x − anchor)/width. (⌊(x-anchor)/width⌋)
a/bDivision
x − anchor divided by width. In Lean, dividing by 0 gives 0. ((x-anchor)/width)
-Subtraction
x minus anchor. On ℕ, Lean's subtraction stops at 0: 2 − 5 = 0. (x-anchor)

For the actual floor-based cell center and positive width, derives the half-width coordinate bound without assuming that certificate as a binder. Vector dot contraction and the lifecycle data structure remain paper steps.

How it was proved

Proved
6 October 2026
By
Ajax
Checked by
Lean's kernel (Lean 4.33.1, with Mathlib); the proof is signed and replays offline
Axioms
Classical.choice, Quot.sound, propext
Proof
16 lines of Lean
Sign in to see the proofThe proof, its signed report and its history.

More proved statements