GameTree.value_Node_ge
theoremverifiedformalizationimported Lean
Key optimality lemma: at a Node m h t, the mover m's coordinate of the backward-induction value dominates every child's.
Statement
set_option autoImplicit true in set_option relaxedAutoImplicit true in variable {N U : Type*} [TotalPreorder U] [DecidableLE U] in theorem GameTree.value_Node_ge (m : N) (h : GameTree N U) (t : List (GameTree N U)) (c : GameTree N U) (hmem : c ∈ h :: t) : (value c) m ≤ (value (Node m h t)) m
- Formalizes
- EconCSLib, `GameTree.value_Node_ge`
- Lean proof
- By xbei (from the file's git history), from github.com/gametheoryinlean/EconCSLib at
1a88f809b538(Copyright (c) 2026 EconCSLib contributors. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.30.0.GameTree.value_Node_ge← GameTree.value_Node_ge (EconCSLib/GameTheory/ExtensiveGame/BackwardInduction.lean:88): verbatim
- Verified
- 08 Oct 2026
- Axioms
Quot.soundpropext- Built on 10
- GameTree.value_LeafGameTree.value_Leaf
- GameTree.value_NodeGameTree.value_Node
- GameTree.valueList_eq_map
valueList lis exactlyl.map value. Useful for rewriting. - GameTree.valueList_nilGameTree.valueList_nil
- GameTree.valueList_consGameTree.valueList_cons
- GameTree.valueGameTree.value
- TotalPreorderA total preorder: a preorder where
≤is total (complete). This is weaker thanLinearOrder— it does NOT require antisymmetry or… - GameTreeA finite extensive-form game of perfect information, without chance. *
Leaf payoff— a terminal state with a payoff vectorN → U. *… - List.argMaxOnA chosen maximizer of
fon the non-empty listhead :: tail, computed by a left fold that keeps the running maximizer (ties keep… - List.argMaxOn_geSoundness: every element's
f-image is≤the chosen maximizer's.
- Used by 3
- GameTree.value₀_Node_one_le_childAt a zero-sum player-1 node,
value₀is no greater than thevalue₀of every child. Equivalently, player 1's local maximization of their… - GameTree.optStrategy_isSubgamePerfectOptimality of
optStrategy: for every subtreeg, every playeri, and everyi-variant deviationσ', the deviating outcome is… - GameTree.value₀_Node_zero_ge_childAt a player-0 node,
value₀is at least thevalue₀of every child.
- GameTree.value₀_Node_one_le_childAt a zero-sum player-1 node,
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 autoImplicit true inset_option relaxedAutoImplicit true invariable {N U : Type*} [TotalPreorder U] [DecidableLE U] in/-- **Key optimality lemma**: at a `Node m h t`, the mover `m`'s coordinate of the backward-induction value dominates every child's. -/theorem GameTree.value_Node_ge (m : N) (h : GameTree N U) (t : List (GameTree N U)) (c : GameTree N U) (hmem : c ∈ h :: t) : (value c) m ≤ (value (Node m h t)) m := by rw [value_Node] -- The argmax is chosen from (value h :: valueList t) w.r.t. `(·) m`. -- We need: `value c m ≤ argMaxOn (·m) (value h) (valueList t) m`. -- Since `c ∈ h :: t`, `value c ∈ value h :: valueList t`. have hcv : value c ∈ value h :: valueList t := by rcases List.mem_cons.mp hmem with rfl | hmem' · exact List.mem_cons_self · refine List.mem_cons.mpr (Or.inr ?_) rw [valueList_eq_map] exact List.mem_map_of_mem hmem' exact List.argMaxOn_ge (fun v => v m) (value h) (valueList t) (value c) hcv