Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§15.5, p. 262 — summed player regrets with smoothness coefficients

Proved
ModernOnlineLearning.SaddlePoint.regret_sum_15_26

by mikedeng1 · Oct 9, 2026 · Mathlib 0df444a (Lean v4.33.1)

online-learningoptimistic-ftrlp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1saddle-point

Let both players run Algorithm 15.6 on nonempty closed convex sets X,YX,YX,Y. Assume fff is convex in xxx and concave in yyy, its partial gradients satisfy (15.7)–(15.10), and the regularizers are respectively λX\lambda_XλX​- and λY\lambda_YλY​-strongly convex, with positive parameters. For every u∈Xu\in Xu∈X and v∈Yv\in Yv∈Y,

∑t=1T[f(xt,v)−f(u,yt)]≤ψX(u)−ψX(x1)+ψY(v)−ψY(y1)+∥gX,1∥X,∗2λX+∥gY,1∥Y,∗2λY+∑t=2T(2LXX2λX+2LXY2λY−λX4)∥xt−xt−1∥X2+∑t=2T(2LYY2λY+2LXY2λX−λY4)∥yt−yt−1∥Y2.\begin{aligned} \sum_{t=1}^T[f(x_t,v)-f(u,y_t)]\le {}&\psi_X(u)-\psi_X(x_1)+\psi_Y(v)-\psi_Y(y_1) +\frac{\|g_{X,1}\|_{X,*}^2}{\lambda_X}+\frac{\|g_{Y,1}\|_{Y,*}^2}{\lambda_Y}\\ &+\sum_{t=2}^T\left(\frac{2L_{XX}^2}{\lambda_X}+\frac{2L_{XY}^2}{\lambda_Y}-\frac{\lambda_X}{4}\right)\|x_t-x_{t-1}\|_X^2\\ &+\sum_{t=2}^T\left(\frac{2L_{YY}^2}{\lambda_Y}+\frac{2L_{XY}^2}{\lambda_X}-\frac{\lambda_Y}{4}\right)\|y_t-y_{t-1}\|_Y^2. \end{aligned}t=1∑T​[f(xt​,v)−f(u,yt​)]≤​ψX​(u)−ψX​(x1​)+ψY​(v)−ψY​(y1​)+λX​∥gX,1​∥X,∗2​​+λY​∥gY,1​∥Y,∗2​​+t=2∑T​(λX​2LXX2​​+λY​2LXY2​​−4λX​​)∥xt​−xt−1​∥X2​+t=2∑T​(λY​2LYY2​​+λX​2LXY2​​−4λY​​)∥yt​−yt−1​∥Y2​.​

This is the intermediate estimate immediately preceding Theorem 15.26; its two stability coefficients determine the theorem's required lower bounds on λX\lambda_XλX​ and λY\lambda_YλY​.

Formalization Note The dual norms are operator norms. The sequence starts at round one, and the stability sums start at round two. The run is constrained only on rounds 1,…,T1,\ldots,T1,…,T; the statement involves no iterate after TTT.

Preamble
import Mathlib
import Definitions.Def_ModernOnlineLearning_SaddlePoint_Setting

set_option autoImplicit false
Formal statement
namespace ModernOnlineLearning.SaddlePoint

/-- The summed-regret display in §15.5, p. 262, immediately before Theorem 15.26. -/
theorem regret_sum_15_26 {EX EY : Type*}
    [NormedAddCommGroup EX] [NormedSpace ℝ EX] [FiniteDimensional ℝ EX]
    [NormedAddCommGroup EY] [NormedSpace ℝ EY] [FiniteDimensional ℝ EY]
    (X : Set EX) (Y : Set EY) (f : EX → EY → ℝ)
    (gx : EX → EY → EX →L[ℝ] ℝ) (gy : EX → EY → EY →L[ℝ] ℝ)
    (ψX : EX → ℝ) (ψY : EY → ℝ) (x : ℕ → EX) (y : ℕ → EY)
    (LXX LXY LYY lamX lamY : ℝ) (T : ℕ)
    (hT : 1 ≤ T) (hXne : X.Nonempty) (hYne : Y.Nonempty)
    (hXclosed : IsClosed X) (hYclosed : IsClosed Y)
    (hX : Convex ℝ X) (hY : Convex ℝ Y)
    (hfx : ∀ v ∈ Y, ConvexOn ℝ X (fun w => f w v))
    (hfy : ∀ u ∈ X, ConcaveOn ℝ Y (f u))
    (hLXX : 0 ≤ LXX) (hLXY : 0 ≤ LXY) (hLYY : 0 ≤ LYY)
    (hlamX : 0 < lamX) (hlamY : 0 < lamY)
    (hψX : StrongConvexOn X lamX ψX) (hψY : StrongConvexOn Y lamY ψY)
    (hsmooth : IsSmoothSaddleOn X Y f gx gy LXX LXY LYY)
    (hrun : IsOptFTRLSaddleRun X Y ψX ψY gx gy x y T) :
    ∀ u ∈ X, ∀ v ∈ Y,
      (∑ t ∈ Finset.Icc 1 T, (f (x t) v - f u (y t))) ≤
      ψX u - ψX (x 1) + ψY v - ψY (y 1) +
      ‖gx (x 1) (y 1)‖ ^ 2 / lamX + ‖gy (x 1) (y 1)‖ ^ 2 / lamY +
      (∑ t ∈ Finset.Icc 2 T,
        ((2 * LXX ^ 2 / lamX + 2 * LXY ^ 2 / lamY - lamX / 4) *
          ‖x t - x (t - 1)‖ ^ 2)) +
      (∑ t ∈ Finset.Icc 2 T,
        ((2 * LYY ^ 2 / lamY + 2 * LXY ^ 2 / lamX - lamY / 4) *
          ‖y t - y (t - 1)‖ ^ 2)) := by sorry

end ModernOnlineLearning.SaddlePoint
Source
Orabona, arXiv:1912.13213v10, §15.5, summed-regret display before Theorem 15.26, p. 262
Human review
  • Endorsed by Shuze Chen · Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 9, 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