Deep linear network: two-point correlator (eqs. 3.12, 3.15)
ProvedDeepLearningTheory.linearNet_two_pointFor a zero-bias deep linear network with positive widths at initialization (eq. 3.4), any two inputs , every layer and neurons of that layer,
The covariance is multiplied by at each layer, which is the origin of the criticality condition .
import Mathlib import Definitions.Def_DLT_DeepLinearNetwork open MeasureTheory ProbabilityTheory open scoped NNReal
namespace DeepLearningTheory
theorem linearNet_two_point {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω)
[IsProbabilityMeasure P] (n : ℕ → ℕ) (hn : ∀ ℓ, 0 < n ℓ) (CW : ℝ≥0)
(W : Ω → ℕ → ℕ → ℕ → ℝ) (hW : IsLinearNetInit P n CW W) (x₁ x₂ : ℕ → ℝ)
(ℓ i₁ i₂ : ℕ) (hℓ : 1 ≤ ℓ) (hi₁ : i₁ < n ℓ) (hi₂ : i₂ < n ℓ) :
∫ ω, linearPreact n (W ω) x₁ ℓ i₁ * linearPreact n (W ω) x₂ ℓ i₂ ∂P
= kron i₁ i₂ * (CW : ℝ) ^ ℓ * inputKernel (n 0) x₁ x₂ := by sorry
end DeepLearningTheoryRead-back
What the Lean code literally says, in plain math · Aristotle (Harmonic) - non-blind, same agent as the drafter
Non-blind read-back. This read-back was written by the same agent that drafted the Lean statement, with full knowledge of the source text and of the intended meaning. It is not independent testimony and must not be mistaken for a blind audit; reviewers should compare the Lean code against the source themselves.
Let be a probability space, widths with for every , and satisfying IsLinearNetInit. For all inputs , every and all :
with the zero-bias linear preactivations.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.