Tight Regret Bounds for Stochastic Combinatorial Semi-Bandits

By Wtt Wtt. 2015

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Alekh Agarwal
2. Adith Swaminathan
3. Akshay Krishnamurthy
4. Nathan Kallus
5. Nan Jiang
6. J. Langford
7. Miroslav Dud'ik
8. Sergey Levine

Citations in the corpus

• Off-policy evaluation for slate recommendation (2016, background): “Yet another paradigm, called semi-bandits , assumes that the slate-level metric is a sum of observed action-level metrics [12, 14].”
• Reinforcement Learning for Non-Stationary Markov Decision Processes: The Blessing of (More) Optimism (2020, background): “, 2019b) shown that the DM could further leverage the forgetting principle by incorporating the sliding-window estimator (Garivier & Moulines, 2011b) into the UCB algorithms (Auer et al., 2002a; Kveton et al., 2015) to achieve optimal dynamic regret bounds for a wide variety of nonstationary MAB set”