LeanMachineLearning

2.9. Online.Bandit.Algorithms.Regret.UCB🔗

From the authors

Regret bounds for the UCB algorithm

Module LeanMachineLearning.Online.Bandit.Algorithms.Regret.UCB contains 18 exposed declarations.

Bandits.UCB.gap_le_two_mul_ucbWidthLemma · 6 deps
Bandits.UCB.pullCount_le_of_ucbIndex_leLemma · 6 deps
Bandits.UCB.prob_ucbIndex_leLemma · 12 deps
Bandits.UCB.prob_lcbIndex_geLemma · 12 deps
Bandits.UCB.pullCount_le_add_threeLemma · 5 deps
Bandits.UCB.pullCount_le_add_three_aeLemma · 26 deps
Bandits.UCB.sum_indicator_good_event_eq_zeroLemma · 27 deps
Bandits.UCB.pullCount_ae_le_add_twoLemma · 27 deps
Bandits.UCB.constSumDefinition · 0 deps
Bandits.UCB.constSum_nonnegLemma · 1 deps
Bandits.UCB.ofReal_constSumLemma · 1 deps
Bandits.UCB.constSum_leLemma · 1 deps
Bandits.UCB.expectation_pullCount_le'Lemma · 24 deps
Bandits.UCB.expectation_pullCount_leLemma · 24 deps
Bandits.UCB.regret_leLemma · 24 deps
Bandits.UCB.regret_le_of_gt_twoLemma · 23 deps
Bandits.UCB.regret_le'Lemma · 24 deps
Bandits.UCB.regret_le_of_gt_two'Theorem · 23 deps