Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Supremum identities for extended-valued superadditive sequences

Definition
CRCD_QuantumChannelContinuity_RegularizationSup

by JWang226 · Oct 7, 2026 · Mathlib c5ea003 (Lean v4.30.0)

channel-renyi-continuity-jwang226quantum-information

Let f:N→[0,∞]f:\mathbb N\to[0,\infty]f:N→[0,∞] satisfy f(m)+f(n)≤f(m+n)f(m)+f(n)\le f(m+n)f(m)+f(n)≤f(m+n) for all natural numbers m,nm,nm,n. Then

nf(k)≤f(nk)(n,k∈N).n f(k)\le f(nk)\quad(n,k\in\mathbb N).nf(k)≤f(nk)(n,k∈N).

Writing S(f)=sup⁡k≥1f(k)/kS(f)=\sup_{k\ge1}f(k)/kS(f)=supk≥1​f(k)/k, every integer n≥1n\ge1n≥1 satisfies

sup⁡k≥1f(nk)k=nS(f).\sup_{k\ge1}\frac{f(nk)}k=nS(f).k≥1sup​kf(nk)​=nS(f).

The denominator on the left is kkk, not nknknk. The same supremum identity holds under the weaker directly supplied repetition premise nf(k)≤f(nk)n f(k)\le f(nk)nf(k)≤f(nk) for all n,kn,kn,k, without requiring full superadditivity. Supremum indices exclude zero, and all arithmetic is in the extended nonnegative reals, allowing f(k)=∞f(k)=\inftyf(k)=∞ and S(f)=∞S(f)=\inftyS(f)=∞.

Definition code
import Mathlib.Data.ENNReal.Inv

/-
Copyright (c) 2026 Jinzhao Wang. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Jinzhao Wang (AI-assisted formalization)
-/



/-!
# Block suprema on multiples

The regularization identity on positive multiples follows directly from
superadditivity. All operations retain infinite values.
-/

open scoped ENNReal
namespace QuantumChannelContinuity

/-- Repeating a block gives the natural-multiple bound for an extended-real
superadditive sequence, including zero and infinite values. -/
theorem ennreal_nat_mul_le_apply_mul (f : ℕ → ℝ≥0∞)
    (hf : ∀ m n, f m + f n ≤ f (m + n)) (n k : ℕ) :
    (n : ℝ≥0∞) * f k ≤ f (n * k) := by
  induction n with
  | zero => simp
  | succ n ih =>
    calc
      ((n + 1 : ℕ) : ℝ≥0∞) * f k = (n : ℝ≥0∞) * f k + f k := by
        simp [add_mul]
      _ ≤ f (n * k) + f k := add_le_add ih le_rfl
      _ ≤ f (n * k + k) := hf _ _
      _ = f ((n + 1) * k) := by rw [Nat.add_mul, one_mul]

/-- Exact positive-block scaling of a regularized supremum, using only the
block repetition inequality. No limit or finiteness premise is needed. -/
theorem ennreal_iSup_div_multiples_of_repeat (f : ℕ → ℝ≥0∞)
    (hf : ∀ n k : ℕ, (n : ℝ≥0∞) * f k ≤ f (n * k))
    {n : ℕ} (hn : 0 < n) :
    (⨆ k : ℕ, ⨆ (_ : 0 < k), f (n * k) / (k : ℝ≥0∞)) =
      (n : ℝ≥0∞) * (⨆ k : ℕ, ⨆ (_ : 0 < k), f k / (k : ℝ≥0∞)) := by
  apply le_antisymm
  · refine iSup_le fun k => iSup_le fun hk => ?_
    have hnk : 0 < n * k := Nat.mul_pos hn hk
    have hbound : f (n * k) / ((n * k : ℕ) : ℝ≥0∞) ≤
        ⨆ j : ℕ, ⨆ (_ : 0 < j), f j / (j : ℝ≥0∞) :=
      le_iSup_of_le (n * k) (le_iSup_of_le hnk le_rfl)
    have hraw := (ENNReal.div_le_iff (by exact_mod_cast hnk.ne')
      (ENNReal.natCast_ne_top (n * k))).mp hbound
    apply (ENNReal.div_le_iff (by exact_mod_cast hk.ne') (by simp)).mpr
    simpa only [Nat.cast_mul, mul_assoc, mul_comm, mul_left_comm] using hraw
  · simp_rw [ENNReal.mul_iSup]
    refine iSup_le fun k => iSup_le fun hk => ?_
    apply le_iSup_of_le k
    apply le_iSup_of_le hk
    simpa only [div_eq_mul_inv, mul_assoc] using
      mul_le_mul_right' (hf n k) (k : ℝ≥0∞)⁻¹

/-- A superadditive sequence has the exact regularization scaling on all
positive block multiples. -/
theorem ennreal_iSup_div_multiples (f : ℕ → ℝ≥0∞)
    (hf : ∀ m n, f m + f n ≤ f (m + n)) {n : ℕ} (hn : 0 < n) :
    (⨆ k : ℕ, ⨆ (_ : 0 < k), f (n * k) / (k : ℝ≥0∞)) =
      (n : ℝ≥0∞) * (⨆ k : ℕ, ⨆ (_ : 0 < k), f k / (k : ℝ≥0∞)) :=
  ennreal_iSup_div_multiples_of_repeat f (ennreal_nat_mul_le_apply_mul f hf) hn

end QuantumChannelContinuity
Source
https://github.com/JWang226/Continuity-of-regularized-channel-renyi-divergence/blob/e29274a69978bb499f00497481c6008befb0b84d/QuantumChannelContinuity/RegularizationSup.lean#L19-L64

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me