prefixHedgeWeight_pos
theoremverifiedimported Lean
Positivity of the prefix Hedge weight. Let $G$ be a finite two-player zero-sum game with $M$ row actions and $N$ column actions, let $\eta \in \bbr$ be a learning rate, and fix any prefix $a_0, \dots, a_{t-1}$ of $t$ column actions together with a row $i$. Then the prefix Hedge weight $w_t(i) = \exp\!\bigl(-\eta \cdot L_t(i)\bigr)$ is strictly positive, where $L_t(i) = \sum_{s=0}^{t-1}\bigl(1 - A_{i,a_s}\bigr)$ is the prefix cumulative loss of row $i$.
Statement
set_option relaxedAutoImplicit false in set_option autoImplicit false in open Real Finset BigOperators in lemma prefixHedgeWeight_pos {M N : ℕ} (G : ZeroSumGame M N) (η : ℝ) {t : ℕ} (actions : Fin t → Fin N) (i : Fin M) : 0 < prefixHedgeWeight G η actions i
- Source
- TCSlib, `prefixHedgeWeight_pos`
- 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.prefixHedgeWeight_pos← prefixHedgeWeight_pos (TCSlib/LearningTheory/Minimax/HedgeInteraction.lean:225): verbatim
- Verified
- 07 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 2
- ZeroSumGameA finite two-player zero-sum game with $M$ row actions and $N$ column actions is a record consisting of a payoff matrix…
- prefixHedgeWeightThe unnormalised Hedge weight of row $i$ after seeing a prefix of column actions at learning rate $\eta$ is…
- Used by 2
- prefixHedgeMixedStrategyThe normalised Hedge distribution after a prefix of $t$ column actions is the mixed row strategy with weights…
- prefixPotential_posPositivity of the prefix potential. Let $G$ be a finite two-player zero-sum game with at least one row action, fix a learning rate…
Lean source view module on GitHub
set_option relaxedAutoImplicit false in set_option autoImplicit false in open Real Finset BigOperators in /-- Every prefix Hedge weight is strictly positive, being an exponential. -/ lemma prefixHedgeWeight_pos {M N : ℕ} (G : ZeroSumGame M N) (η : ℝ) {t : ℕ} (actions : Fin t → Fin N) (i : Fin M) : 0 < prefixHedgeWeight G η actions i := Real.exp_pos _