MLP: second-layer two-point correlator (eq. 4.41)
ProvedDeepLearningTheory.mlp_second_layer_two_pointFor an MLP at initialization with a measurable activation of polynomial growth and , two inputs and second-layer neurons ,
This is the first step of the kernel recursion .
import Mathlib import Definitions.Def_DLT_DeepLinearNetwork import Definitions.Def_DLT_MLPInit open MeasureTheory ProbabilityTheory open scoped NNReal
namespace DeepLearningTheory
theorem mlp_second_layer_two_point {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω)
[IsProbabilityMeasure P] (n : ℕ → ℕ) (Cb CW : ℕ → ℝ≥0) (σ : ℝ → ℝ)
(b : Ω → ℕ → ℕ → ℝ) (W : Ω → ℕ → ℕ → ℕ → ℝ) (hinit : IsMLPInit P n Cb CW b W)
(hσm : Measurable σ) (hσ : HasPolyGrowth σ) (hn₁ : 0 < n 1) (x : Fin 2 → ℕ → ℝ) (i₁ i₂ : ℕ) (hi₁ : i₁ < n 2) (hi₂ : i₂ < n 2) :
∫ ω, mlpPreact n σ (b ω) (W ω) (x 0) 2 i₁ * mlpPreact n σ (b ω) (W ω) (x 1) 2 i₂ ∂P
= kron i₁ i₂ * ((Cb 2 : ℝ) + (CW 2 : ℝ) *
gaussAvg (firstLayerMetric (Cb 1) (CW 1) (n 0) x) (fun u => σ (u 0) * σ (u 1))) := 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, , any function, and random bias/weight fields satisfying IsMLPInit (all biases and weights with , , mutually independent; , ). Write for the MLP preactivation at and for the first-layer metric of the listed inputs. Assume moreover that is measurable and has polynomial growth (). Assume . Then for every pair of inputs and all : , with the first-layer metric of and . Here is the integral of against Mathlib's centered multivariate Gaussian with covariance matrix (the inputs may repeat, so may be singular).
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.