OAI.SphericalPerceptron.finiteSphericalFieldValue_tendsto_dual: intermediate lemma from OpenAI, Perceptron free energies and microscopic jamming exponents (result family 222)

Proved

OAI.SphericalPerceptron.finiteSphericalFieldValue_tendsto_dual

by dbenbenn · Oct 25, 2026 · Mathlib 0df444a (Lean v4.33.1)

openai-math

An intermediate lemma of the split of OpenAI's proof of OAI.SphericalPerceptron.spherical_linear_field_formula, carved from OpenAI's openai/math development for this platform. The statement and its proof are OpenAI's (OpenAI, Perceptron free energies and microscopic jamming exponents (result family 222), OpenAI Math Release, September 24, 2026, result family 222, github.com/openai/math, Apache License 2.0); in Lean it is the declaration OAI.SphericalPerceptron.finiteSphericalFieldValue_tendsto_dual of lean/OAI/Probability/SphericalField/Fields/DepthUniform.lean at commit adc7f12, ported to Lean 4.33.1 and Mathlib 0df444a.

The statement, as published (namespace OAI.SphericalPerceptron, with open MeasureTheory ProbabilityTheory Filter Set, open scoped Topology NNReal ENNReal BigOperators):

lemma finiteSphericalFieldValue_tendsto_dual {k : ℕ} (w h : Fin (k+1) → ℝ)
    (hw : ∀ i, 0 < w i) (hw1 : ∑ i, w i=1) (hh : Monotone h) (hh0 : 0 ≤ h 0) :
    Tendsto (fun n : ℕ => finiteSphericalFieldValue n k w h) atTop
      (𝓝 (sInf (finiteSphericalDualValues w h (fun i => (hw i).le) hw1)))

Definitions. The statement is written with OAI.SphericalPerceptron.finiteSphericalDualValues and OAI.SphericalPerceptron.finiteSphericalFieldValue (the published bundle Def_SphericalField).

Used by. The proof of OAI.SphericalPerceptron.finiteSphericalFieldValue_all_depths_uniform of this split imports it.

Preamble

import Mathlib
import Definitions.Def_SphericalField

Formal statement

section

set_option linter.all false
set_option linter.unusedVariables false
set_option linter.unusedTactic false
set_option linter.unreachableTactic false

namespace OAI

section
noncomputable section
open MeasureTheory ProbabilityTheory Filter Set
open scoped Topology NNReal ENNReal BigOperators

namespace SphericalPerceptron

lemma finiteSphericalFieldValue_tendsto_dual {k : ℕ} (w h : Fin (k+1) → ℝ)
    (hw : ∀ i, 0 < w i) (hw1 : ∑ i, w i=1) (hh : Monotone h) (hh0 : 0 ≤ h 0) :
    Tendsto (fun n : ℕ => finiteSphericalFieldValue n k w h) atTop
      (𝓝 (sInf (finiteSphericalDualValues w h (fun i => (hw i).le) hw1))) := by
  sorry

end SphericalPerceptron
end
end

end OAI
end

Source

OpenAI, Perceptron free energies and microscopic jamming exponents (result family 222), OpenAI Math Release, September 24, 2026, https://github.com/openai/math (result family 222; paper: https://github.com/openai/math/blob/adc7f1241b42e322a6451854ab7e4b4c146bf78a/lean/ComparatorChallenges/SphericalField.lean; Apache License 2.0); Lean: https://github.com/openai/math/blob/adc7f1241b42e322a6451854ab7e4b4c146bf78a/lean/OAI/Probability/SphericalField/Fields/DepthUniform.lean (declaration OAI.SphericalPerceptron.finiteSphericalFieldValue_tendsto_dual)

View graph