Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 1 — strong duality for the worst-case expected loss over an optimal-transport ball, with the minimum attained

Proved
RWPI.SqrtLasso.proposition_1

by mikedeng1 · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

distributionally-robust-optimizationdualityoptimal-transportp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Let (X1,Y1),…,(Xn,Yn)∈Rd×R(X_1,Y_1),\dots,(X_n,Y_n)\in\mathbb R^d\times\mathbb R(X1​,Y1​),…,(Xn​,Yn​)∈Rd×R be training data, n≥1n\ge1n≥1, with empirical distribution Pn=1n∑i=1nδ(Xi,Yi)P_n = \frac1n\sum_{i=1}^n\delta_{(X_i,Y_i)}Pn​=n1​∑i=1n​δ(Xi​,Yi​)​. Let c:Rd+1×Rd+1→[0,∞]c:\mathbb R^{d+1}\times\mathbb R^{d+1}\to[0,\infty]c:Rd+1×Rd+1→[0,∞] be a lower semicontinuous cost with c(z,z)=0c(z,z) = 0c(z,z)=0 for every zzz, and let l:Rd×R→[0,∞)l:\mathbb R^d\times\mathbb R\to[0,\infty)l:Rd×R→[0,∞) be an upper semicontinuous loss. For γ≥0\gamma\ge0γ≥0 let

φγ(Xi,Yi)=sup⁡u∈Rd, v∈R{l(u,v)−γ c((u,v),(Xi,Yi))}(11)\varphi_\gamma(X_i,Y_i) = \sup_{u\in\mathbb R^d,\ v\in\mathbb R}\big\{ l(u,v) - \gamma\, c\big((u,v),(X_i,Y_i)\big)\big\} \qquad (11)φγ​(Xi​,Yi​)=u∈Rd, v∈Rsup​{l(u,v)−γc((u,v),(Xi​,Yi​))}(11)

(terms with infinite cost do not contribute). Then for every δ>0\delta>0δ>0

sup⁡P: Dc(P,Pn)≤δEP[l(X,Y)]=min⁡γ≥0{γδ+1n∑i=1nφγ(Xi,Yi)},\sup_{P:\ D_c(P,P_n)\le\delta} \mathbb E_P\big[l(X,Y)\big] = \min_{\gamma\ge0}\Big\{ \gamma\delta + \frac1n\sum_{i=1}^n \varphi_\gamma(X_i,Y_i)\Big\},P: Dc​(P,Pn​)≤δsup​EP​[l(X,Y)]=γ≥0min​{γδ+n1​i=1∑n​φγ​(Xi​,Yi​)},

and the minimum is attained: there is γ∗≥0\gamma^\ast\ge0γ∗≥0 achieving equality, and the left side is at most the bracket for every γ≥0\gamma\ge0γ≥0.

The proposition turns the supremum over uncountably many probability measures into a one-dimensional convex minimization, in which φγ\varphi_\gammaφγ​ is often explicit. The paper quotes it from Blanchet and Murthy (Math. Oper. Res., 2019, Theorem 1). Taking the infimum over the regression parameter on both sides gives the paper's identity (12); in the paper the loss is l(⋅;β)l(\cdot;\beta)l(⋅;β), and the statement here holds for each fixed β\betaβ.

Formalization Note Two hypotheses are added to the page and recorded: l≥0l\ge0l≥0 (every loss in the paper is nonnegative; the expectation is a lower Lebesgue integral), and δ>0\delta>0δ>0 (assumed in Blanchet–Murthy; at δ=0\delta = 0δ=0 the minimum need not be attained). Both sides take values in [0,∞][0,\infty][0,∞]; "min" is stated as attainment plus a lower bound. Identity (12) is not stated separately. The data space is (Fin d→R)×R(\text{Fin } d\to\mathbb R)\times\mathbb R(Fin d→R)×R with the product topology.

Preamble
import Mathlib
import Definitions.Def_WassersteinDRO_Regularization_empiricalDistribution
import Definitions.Def_RWPI_SqrtLasso_worstCase
import Definitions.Def_RWPI_SqrtLasso_phi

open MeasureTheory
Formal statement
namespace RWPI.SqrtLasso

/-- Proposition 1 (Blanchet, Kang & Murthy, p. 10; quoted from Blanchet & Murthy, Math. Oper. Res.
2019, Theorem 1), first identity: strong duality for the worst-case expected loss over the
optimal-transport ball `{P : D_c(P, P_n) ≤ δ}` around the empirical distribution `P_n` of the data
`(X_i, Y_i)`, with the minimum over `γ ≥ 0` attained.
The cost `c` is lower semicontinuous, `[0, ∞]`-valued and vanishes on the diagonal; the loss `l` is
upper semicontinuous. Added hypotheses (recorded): `l ≥ 0` (every loss of the paper) and `δ > 0`
(assumed in Blanchet–Murthy; at `δ = 0` the minimum need not be attained). -/
theorem proposition_1 {d n : ℕ} (hn : 0 < n) (X : Fin n → Fin d → ℝ) (Y : Fin n → ℝ)
    (c : (Fin d → ℝ) × ℝ → (Fin d → ℝ) × ℝ → ENNReal)
    (hc : LowerSemicontinuous (Function.uncurry c)) (hc0 : ∀ z, c z z = 0)
    (l : (Fin d → ℝ) × ℝ → ℝ) (hl : UpperSemicontinuous l) (hl0 : ∀ z, 0 ≤ l z)
    (δ : ℝ) (hδ : 0 < δ) :
    ∃ γ : ℝ, 0 ≤ γ ∧
      worstCase c δ (WassersteinDRO.Regularization.empiricalDistribution
          (fun i : Fin n => (X i, Y i))) l =
        ENNReal.ofReal (γ * δ) + (n : ENNReal)⁻¹ * ∑ i, phi c l γ (X i, Y i) ∧
      ∀ γ' : ℝ, 0 ≤ γ' →
        worstCase c δ (WassersteinDRO.Regularization.empiricalDistribution
            (fun i : Fin n => (X i, Y i))) l ≤
          ENNReal.ofReal (γ' * δ) + (n : ENNReal)⁻¹ * ∑ i, phi c l γ' (X i, Y i) := by sorry

end RWPI.SqrtLasso
Source
Blanchet, Kang & Murthy, Robust Wasserstein Profile Inference and Applications to Machine Learning, arXiv:1610.05627v4, p. 10, Proposition 1 (first identity; quoted from Blanchet & Murthy 2019, Theorem 1)
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me