`BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread` {beta D : ℝ} (hb : 0 ≤ beta) (s : Fin m → ℝ) (j : Fin m) (hD : ∀ l, s l ≤ s j + D) : Real.exp (-(beta * D)) / (m : ℝ

Proved

BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread

by leonardopedro · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

theoremstimepiece

Prove the following Lean 4 theorem from ChapterAttentionRetrieval.

BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread {beta D : ℝ} (hb : 0 ≤ beta) (s : Fin m → ℝ) (j : Fin m) (hD : ∀ l, s l ≤ s j + D) : Real.exp (-(beta * D)) / (m : ℝ) ≤ scoreSoftmax beta s j

Formalization note: Lean 4 identifier BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread.

Preamble

-- Generated from ChapterAttentionRetrieval.lean — theorem BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread
import Definitions.Def_ChapterSoftmaxOrder
import Mathlib
import Definitions.Def_ChapterAttentionRetrieval
import Definitions.Def_ChapterSoftmaxSharpness
open BookProof.ChapterSoftmaxSharpness
open BookProof.ChapterAttentionRetrieval


open scoped BigOperators

noncomputable section


open BookProof.ChapterSoftmaxSharpness BookProof.ChapterSoftmaxOrder

variable {m n : ℕ} {E : Type*} [NormedAddCommGroup E] [NormedSpace ℝ E]

Formal statement

theorem BookProof.ChapterAttentionRetrieval.scoreSoftmax_ge_of_spread {beta D : ℝ} (hb : 0 ≤ beta) (s : Fin m → ℝ)
    (j : Fin m) (hD : ∀ l, s l ≤ s j + D) :
    Real.exp (-(beta * D)) / (m : ℝ) ≤ scoreSoftmax beta s j := by sorry

View graph