2.4. Online.Bandit.BayesRegret
Bayesian regret
This file defines actionMean, bestAction, gap, and regret as random variables in a
measurable space Ω. These definitions are useful when IsBayesAlgEnvSeq Q κ alg E A Y P.
Recall that IsBayesAlgEnvSeq Q κ alg E A Y P states that there is a measure P : Measure Ω such
that the parameter E : Ω → 𝓔 has law Q and that the sequences of actions A : ℕ → Ω → 𝓐 and
feedbacks Y : ℕ → Ω → 𝓨 are generated by the algorithm alg : Algorithm 𝓐 𝓨 interacting with an
underlying environment that depends on E and κ (stationaryEnv (κ.sectR (E ω)))
Main definitions
-
actionMean κ E a: the mean feedback associated with actiona : 𝓐based on the parameterE, which defines the underlying stationary environment together with the kernelκ. -
bestAction κ E: (one of) the action(s) with the highest associated mean feedback based onE. -
gap κ E A n: the difference between the highest mean feedback associated with an action and the mean feedback associated with the action at timenbased onEand the sequence of actionsA. -
regret κ E A n: the regret at timenbased onEand the sequence of actionsA. IfIsBayesAlgEnvSeq Q κ alg E A Y P, thenP[regret κ E A n]is the so-called Bayesian regret of algorithmalgunder the priorQ.
Module LeanMachineLearning.Online.Bandit.BayesRegret contains 17 exposed declarations.
-
Learning.IsBayesAlgEnvSeq.actionMean -
Learning.IsBayesAlgEnvSeq.measurable_actionMean -
Learning.IsBayesAlgEnvSeq.measurable_uncurry_actionMean_comp -
Learning.IsBayesAlgEnvSeq.integrable_uncurry_actionMean_comp -
Learning.IsBayesAlgEnvSeq.bestAction -
Learning.IsBayesAlgEnvSeq.measurable_bestAction -
Learning.IsBayesAlgEnvSeq.gap -
Learning.IsBayesAlgEnvSeq.gap_nonneg_of_le -
Learning.IsBayesAlgEnvSeq.gap_le_of_mem_Icc -
Learning.IsBayesAlgEnvSeq.gap_eq_sub -
Learning.IsBayesAlgEnvSeq.measurable_gap -
Learning.IsBayesAlgEnvSeq.integrable_gap -
Learning.IsBayesAlgEnvSeq.regret -
Learning.IsBayesAlgEnvSeq.regret_eq_sum_gap -
Learning.IsBayesAlgEnvSeq.regret_eq_sum_gap' -
Learning.IsBayesAlgEnvSeq.measurable_regret -
Learning.IsBayesAlgEnvSeq.integrable_regret