BanditRLlib
Lean gate passed before this site build; local proof declarations are shown as compiled.

Lean module · Tsallis-FTRL

BanditRLProof.TsallisRegularizer

# Tsallis finite-simplex regularizer wrappers This module records the first deterministic `Real.rpow` surface for Tsallis-INF/FTRL routes. It defines the finite-action Tsallis power sum, Tsallis entropy, and the negative-entropy regularizer used as an FTRL regularizer, then packages the small well-definedness side conditions over the finite simplex. It does not prove convexity, stability/penalty bounds, self-bounding conversion, learning-rate optimization, or regret.

Module map

Declarations
6
Placeholders
0

Imports

BanditRLProof.FTRLOneStep

Imported by

BanditRLProof, BanditRLProof.TsallisFTRLRegret, BanditRLProof.TsallisImportanceWeightedMoment

Declarations

Open an item to read its exact compact statement and source link. Detailed teaching notes are linked when registered.

def BanditRLProof.Tsallis.powerSum Compiled

Finite Tsallis power sum `sum_a p_a^alpha`.

noncomputable def powerSum {Action : Type u} (arms : Finset Action) (alpha : Real) (p : Action -> Real) : Real
def BanditRLProof.Tsallis.entropy Compiled

Tsallis entropy on a finite action set, with denominator left explicit.

noncomputable def entropy {Action : Type u} (arms : Finset Action) (alpha : Real) (p : Action -> Real) : Real
def BanditRLProof.Tsallis.negEntropyRegularizer Compiled

Negative Tsallis entropy as the FTRL regularizer convention.

noncomputable def negEntropyRegularizer {Action : Type u} (arms : Finset Action) (alpha : Real) (p : Action -> Real) : Real
theorem BanditRLProof.Tsallis.one_sub_exponent_ne_zero Compiled

The Tsallis denominator is nonzero when `alpha != 1`.

theorem one_sub_exponent_ne_zero {alpha : Real} (halpha : alpha ≠ 1) : 1 - alpha ≠ 0
theorem BanditRLProof.Tsallis.powerSum_nonneg_of_finiteSimplex Compiled

The Tsallis power sum is nonnegative on the finite simplex.

theorem powerSum_nonneg_of_finiteSimplex {Action : Type u} (arms : Finset Action) (alpha : Real) (p : Action -> Real) (hp : FTRL.finiteSimplex arms p) : 0 <= powerSum arms alpha p
theorem BanditRLProof.Tsallis.negEntropyRegularizer_wellDefined_on_finiteSimplex Compiled

Well-definedness package for the finite-simplex Tsallis regularizer. The two facts exposed here are the local obligations needed before later Tsallis/FTRL leaves can use `Real.rpow` algebra and division by `1 - alpha`.

theorem negEntropyRegularizer_wellDefined_on_finiteSimplex {Action : Type u} (arms : Finset Action) (alpha : Real) (p : Action -> Real) (hp : FTRL.finiteSimplex arms p) (halpha : alpha ≠ 1) : 0 <= powerSum arms alpha p ∧ 1 - alpha ≠ 0 ∧ negEntropyRegularizer arms alpha p = - ((powerSum arms alpha p - 1) / (1 - alpha))