Nearly Optimal Algorithms for Linear Contextual Bandits with Adversarial Corruptions.
Jiafan HeDongruo ZhouTong ZhangQuanquan GuPublished in: CoRR (2022)
Keyphrases
- data structure
- worst case
- optimal linear
- computationally efficient
- learning algorithm
- regret bounds
- machine learning algorithms
- orders of magnitude
- times faster
- computational efficiency
- benchmark datasets
- approximately optimal
- optimality criterion
- exhaustive search
- closed form
- data mining algorithms
- data sets
- optimization problems
- dynamic programming
- np hard
- machine learning