§6.2, p. 3608 — W_{2,T}((1/n) Σ δ_{x_i}, (1/n) Σ δ_{y_i})² ≤ (1/n) Σ ‖x_i − y_i‖²_{*,T}
ProvedGraphonMF.DenseLLN.coupling_boundempirical-measurep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1wasserstein-distance
Let and let and be paths in . Then
In the proof of Theorem 3.1 it is applied pathwise with and . It bounds the distance between the two empirical measures by the mean-square error of Lemma 6.1.
Formalization Note The page states the bound for the random empirical measures. It is a deterministic statement about two -tuples of paths, and it is stated in that form (via the coupling ).
Preamble
import Mathlib import Definitions.Def_GraphonMF_DenseLLN_Setting open MeasureTheory ProbabilityTheory Filter Topology open scoped ENNReal NNReal
Formal statement
namespace GraphonMF.DenseLLN
theorem coupling_bound {T : ℝ≥0} {d : ℕ} (n : ℕ) (hn : 0 < n) (x y : Fin n → GraphonMF.Stability.Cd T d) :
GraphonMF.Stability.W2T (empiricalMeasure x) (empiricalMeasure y) ^ 2 ≤
(1 / (n : ℝ≥0∞)) * ∑ i : Fin n, ‖x i - y i‖ₑ ^ 2 := by sorry
end GraphonMF.DenseLLN
Source
Bayraktar, Chakraborty, Wu, Graphon mean field systems, Ann. Appl. Probab. 33(5) (2023), p. 3608, §6.2, display after (6.13)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.