Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

p. 36 — v^DLP − v^FR ≤ Σ_l r^l_max K_l (2√T + √2) + n r_max (log T + 1) + n r_max λ_max

Proved
ResolvingNRM.FRUpper.fr_dlp_gap_explicit

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

network-revenue-managementp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1re-solvingregret-bound

Consider the network revenue management model with Poisson rates λj>0\lambda_j > 0λj​>0, revenues rj≥0r_j \ge 0rj​≥0, nonnegative consumption matrix A=(alj)A = (a_{lj})A=(alj​), initial capacity C≥0C \ge 0C≥0, and an integer horizon T≥1T \ge 1T≥1. Let vDLP=T v(C/T)v^{\mathrm{DLP}} = T\, v(C/T)vDLP=Tv(C/T) be the DLP value and vFRv^{\mathrm{FR}}vFR the expected revenue of the FR policy (Algorithm 2) with any optimal-solution selector. With rmax⁡l=max⁡j{rjI(alj>0)/alj}r^l_{\max} = \max_j \{ r_j \mathbb{I}(a_{lj} > 0)/a_{lj} \}rmaxl​=maxj​{rj​I(alj​>0)/alj​}, Kl=∑jalj2λjK_l = \sqrt{\sum_j a_{lj}^2 \lambda_j}Kl​=∑j​alj2​λj​​, rmax⁡=max⁡jrjr_{\max} = \max_j r_jrmax​=maxj​rj​ and λmax⁡=max⁡jλj\lambda_{\max} = \max_j \lambda_jλmax​=maxj​λj​,

vDLP−vFR ≤ ∑l=1mrmax⁡lKl(2T+2)+n rmax⁡(log⁡T+1)+n rmax⁡λmax⁡.v^{\mathrm{DLP}} - v^{\mathrm{FR}} \ \le\ \sum_{l=1}^m r^l_{\max} K_l \big(2\sqrt{T} + \sqrt{2}\big) + n\, r_{\max} (\log T + 1) + n\, r_{\max} \lambda_{\max} .vDLP−vFR ≤ l=1∑m​rmaxl​Kl​(2T​+2​)+nrmax​(logT+1)+nrmax​λmax​.

This is the explicit form of Proposition 3: every constant depends on λ\lambdaλ, rrr and AAA only, and none on the capacity CCC.

Formalization Note. The bound is the last display of the proof of Proposition 3 (p. 36) with the constant KlK_lKl​ of Lemma 8 corrected from ∑jalj2λj2\sqrt{\sum_j a_{lj}^2\lambda_j^2}∑j​alj2​λj2​​ to ∑jalj2λj\sqrt{\sum_j a_{lj}^2\lambda_j}∑j​alj2​λj​​ (see the Lemma 8 item; the printed constant is false). The printed sum over l=1,…,Ll = 1, \dots, Ll=1,…,L is read as l∈[m]l \in [m]l∈[m]. log⁡\loglog is the natural logarithm. λj>0\lambda_j > 0λj​>0, r≥0r \ge 0r≥0, alj≥0a_{lj} \ge 0alj​≥0, C≥0C \ge 0C≥0 are the model's standing assumptions (p. 7).

Preamble
import Mathlib
import Definitions.Def_RLPBidPrice_Unbiased_Model
import Definitions.Def_ResolvingNRM_FRUpper_Model

open RLPBidPrice.Unbiased Matrix
Formal statement
namespace ResolvingNRM.FRUpper

/-- The explicit bound at the end of the proof of Proposition 3, App. C.2, p. 36, with the
corrected Lemma 8 constant `K_l = √(∑_j a_lj² λ_j)`:
`v^DLP − v^FR ≤ ∑_l r^l_max K_l (2√T + √2) + n r_max (log T + 1) + n r_max λ_max`. -/
theorem fr_dlp_gap_explicit {m n : ℕ} (A : Matrix (Fin m) (Fin n) ℝ) (r lam : Fin n → ℝ)
    (hlam : ∀ j, 0 < lam j) (hr : 0 ≤ r) (hA : ∀ l j, 0 ≤ A l j)
    (sel : (Fin m → ℝ) → (Fin n → ℝ)) (hsel : ResolvingNRM.IRT.IsDLPSelector A r lam sel)
    (T : ℕ) (hT : 1 ≤ T) (C : Fin m → ℝ) (hC : 0 ≤ C) :
    ResolvingNRM.IRT.dlpValue A r lam T C - frValue A r lam sel T C
      ≤ (∑ l, rmaxRes A r l * Kres A lam l) * (2 * Real.sqrt T + Real.sqrt 2)
        + n * rMax r * (Real.log T + 1) + n * rMax r * lamMax lam := by sorry

end ResolvingNRM.FRUpper
Source
Bumpensanti, Wang, A Re-solving Heuristic with Uniformly Bounded Loss for Network Revenue Management, arXiv:1802.06192v3, App. C.2, final display of the proof of Proposition 3, p. 36
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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