Regret bounds for sleeping experts and bandits

DOI10.1007/s10994-010-5178-7zbMath1370.68254MaRDI QIDQ1959599

Alexandru Niculescu-Mizil, Yogeshwer Sharma, Robert D. Kleinberg

Publication date: 7 October 2010

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-010-5178-7

zbMATH Keywords

computational learning theory; online algorithms; regret

Mathematics Subject Classification ID

68T05: Learning and adaptive systems in artificial intelligence

68W27: Online algorithms; streaming algorithms

Related Items

Finite-Time Analysis for the Knowledge-Gradient Policy, Unnamed Item, Near-Optimal Algorithms for Online Matrix Prediction, The \(K\)-armed dueling bandits problem, Sleeping experts and bandits approach to constrained Markov decision processes, Ballooning multi-armed bandits, Truthful Mechanisms with Implicit Payment Computation, Learning Hurdles for Sleeping Experts, Online Collaborative Filtering on Graphs

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Asymptotically efficient adaptive allocation rules
• A game of prediction with expert advice
• The weighted majority algorithm
• Adaptive game playing using multiplicative weights
• Weighted sums of certain dependent random variables
• Efficient algorithms for online decision problems
• How to use expert advice
• The Nonstochastic Multiarmed Bandit Problem
• Probability Inequalities for Sums of Bounded Random Variables
• Learning Theory
• Some aspects of the sequential design of experiments
• Finite-time analysis of the multiarmed bandit problem