Lean module · Foundations
BanditRLProof.Algorithms.HeavyTailUCB
Causal robust UCB with the audited conservative confidence schedule. Time is zero-based; the logarithm uses max(t,2), and failure probability t^-4 is the explicit repair recorded in DERIVATION.md. No regret endpoint is claimed here.
Module map
Imports
BanditRLProof.Algorithms.HeavyTailHistory
Imported by
Declarations
Open an item to read its exact compact statement and source link. Detailed teaching notes are linked when registered.
def
BanditRLProof.HeavyTail.confidenceLog
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.confidenceLogReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def confidenceLog (t : ℕ) : ℝ
def
BanditRLProof.HeavyTail.sampleThreshold
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.sampleThresholdReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def sampleThreshold (ε u : ℝ) (t s : ℕ) : ℝ
def
BanditRLProof.HeavyTail.confidenceRadius
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.confidenceRadiusReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def confidenceRadius (ε u : ℝ) (t count : ℕ) : ℝ
def
BanditRLProof.HeavyTail.historyIndex
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.historyIndexReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def historyIndex (initial : Fin K) (ε u : ℝ) (n : ℕ) (h : History.FinitePairHistory (Fin K) ℝ n) (arm : Fin K) : ℝ
def
BanditRLProof.HeavyTail.nextArm
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.nextArmReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def nextArm (hK : 0 < K) (ε u : ℝ) (n : ℕ) (h : History.FinitePairHistory (Fin K) ℝ n) : Fin K
theorem
BanditRLProof.HeavyTail.measurable_historyIndex
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.measurable_historyIndexReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
theorem measurable_historyIndex (initial : Fin K) (ε u : ℝ) (n : ℕ) (arm : Fin K) : Measurable (fun h => historyIndex initial ε u n h arm)
theorem
BanditRLProof.HeavyTail.measurable_nextArm
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.measurable_nextArmReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
theorem measurable_nextArm (hK : 0 < K) (ε u : ℝ) (n : ℕ) : Measurable (nextArm hK ε u n)
def
BanditRLProof.HeavyTail.robustAction
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.robustActionReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def robustAction (hK : 0 < K) (ε u : ℝ)
def
BanditRLProof.HeavyTail.robustReward
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.robustRewardReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
noncomputable def robustReward (hK : 0 < K) (ε u : ℝ)
theorem
BanditRLProof.HeavyTail.measurable_robustAction
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.measurable_robustActionReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
theorem measurable_robustAction (hK : 0 < K) (ε u : ℝ) (t : ℕ) : Measurable (fun stream => robustAction hK ε u stream t)
theorem
BanditRLProof.HeavyTail.robustAction_initialization
Compiled
No declaration docstring is present; use the chapter context and exact statement below.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.robustAction_initializationReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
theorem robustAction_initialization (hK : 0 < K) (ε u : ℝ) (stream : UCB.ArmRewardStream K) (t : ℕ) (ht : t < K) : robustAction hK ε u stream t = UCB.initializationArm hK t
theorem
BanditRLProof.HeavyTail.robustAction_maximizes
Compiled
The selected index is maximal on the actual observed history.
Used in these reading views: Bandit Book
1. Finite bandits, traces, and regret
Canonical node identity
declaration:BanditRLProof.HeavyTail.robustAction_maximizesReading membership is not a proof dependency. Exact assumptions remain in the Lean statement.
theorem robustAction_maximizes (hK : 0 < K) (ε u : ℝ) (stream : UCB.ArmRewardStream K) (n : ℕ) (hn : K ≤ n+1) (arm : Fin K) : let h := ArmStreamPolicy.history (UCB.initializationArm hK 0) (nextArm hK ε u) stream n historyIndex (UCB.initializationArm hK 0) ε u n h arm ≤ historyIndex (UCB.initializationArm hK 0) ε u n h (robustAction hK ε u stream (n+1))