MLP: (eq. 4.27)
ProvedDeepLearningTheory.mlp_first_layer_activation_two_pointFor an MLP at initialization with a measurable activation of polynomial growth, two inputs and a first-layer neuron ,
where is the expectation over a centered Gaussian vector with covariance given by the first-layer metric, and .
import Mathlib import Definitions.Def_DLT_DeepLinearNetwork import Definitions.Def_DLT_MLPInit open MeasureTheory ProbabilityTheory open scoped NNReal
namespace DeepLearningTheory
theorem mlp_first_layer_activation_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) (x : Fin 2 → ℕ → ℝ) (j : ℕ) (hj : j < n 1) :
∫ ω, σ (mlpPreact n σ (b ω) (W ω) (x 0) 1 j) * σ (mlpPreact n σ (b ω) (W ω) (x 1) 1 j) ∂P
= 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 (). Then for every pair of inputs and every : , with the first-layer metric of . 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.