OnlineLearning.finiteUpperImage_upper
theoremverifiedimported Lean
Finite upper image is upward closed. Let $X, Y \subseteq \bbr$, let $f : \bbr \to \bbr \to \bbr$ be a payoff, and let $u$ be a finite sample of columns from $Y$. If a vector $z : u \to \bbr$ belongs to the finite upper image $U(X,f,u)$, and $z' : u \to \bbr$ satisfies $z_y \le z'_y$ for every $y \in u$, then $z'$ also belongs to $U(X,f,u)$.
Statement
set_option relaxedAutoImplicit false in set_option autoImplicit false in open Real Finset BigOperators in lemma OnlineLearning.finiteUpperImage_upper {X Y : Set ℝ} {f : ℝ → ℝ → ℝ} {u : Finset Y} {z : u → ℝ} (hz : z ∈ finiteUpperImage X f u) {z' : u → ℝ} (hzz' : ∀ y : u, z y ≤ z' y) : z' ∈ finiteUpperImage X f u
- Source
- TCSlib, `OnlineLearning.finiteUpperImage_upper`
- Lean proof
- By Karim Abdel Sadek, Mark Bedaywi, from github.com/Shilun-Allan-Li/tcslib at
c8a02591b1e5(Copyright (c) 2026 Karim Abdel Sadek and Mark Bedaywi. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.25.0.OnlineLearning.finiteUpperImage_upper← OnlineLearning.finiteUpperImage_upper (TCSlib/LearningTheory/Minimax/ConvexMinimaxSeparation.lean:121): verbatim
- Verified
- 07 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 1
- OnlineLearning.finiteUpperImageFor sets $X, Y \subseteq \bbr$, a payoff $f : \bbr \to \bbr \to \bbr$, and a finite sample $u$ of columns from $Y$, the finite upper image…
- Used by 1
- OnlineLearning.separating_coordinate_nonposSeparating functional is coordinatewise nonpositive. Let $X, Y \subseteq \bbr$, let $f : \bbr \to \bbr \to \bbr$ be a payoff, let $u$ be a…
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
set_option relaxedAutoImplicit false inset_option autoImplicit false inopen Real Finset BigOperators in/-- The upper image is coordinatewise upward closed: increasing any coordinates of a member keeps it in the upper image, with the same witness row. -/lemma OnlineLearning.finiteUpperImage_upper {X Y : Set ℝ} {f : ℝ → ℝ → ℝ} {u : Finset Y} {z : u → ℝ} (hz : z ∈ finiteUpperImage X f u) {z' : u → ℝ} (hzz' : ∀ y : u, z y ≤ z' y) : z' ∈ finiteUpperImage X f u := by rcases hz with ⟨x, hx, hz⟩ exact ⟨x, hx, fun y => (hz y).trans_le (hzz' y)⟩