Corpus ID: 219965651

An Empirical Process Approach to the Union Bound: Practical Algorithms for Combinatorial and Linear Bandits

@article{KatzSamuels2020AnEP,
  title={An Empirical Process Approach to the Union Bound: Practical Algorithms for Combinatorial and Linear Bandits},
  author={Julian Katz-Samuels and Lalit Jain and Zohar S. Karnin and Kevin G. Jamieson},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.11685}
}
This paper proposes near-optimal algorithms for the pure-exploration linear bandit problem in the fixed confidence and fixed budget settings. Leveraging ideas from the theory of suprema of empirical processes, we provide an algorithm whose sample complexity scales with the geometry of the instance and avoids an explicit union bound over the number of arms. Unlike previous approaches which sample based on minimizing a worst-case variance (e.g. G-optimal design), we define an experimental design… Expand
Fixed-Budget Best-Arm Identification in Contextual Bandits: A Static-Adaptive Algorithm
Pure Exploration with Structured Preference Feedback

References

SHOWING 1-10 OF 35 REFERENCES
Nearly Instance Optimal Sample Complexity Bounds for Top-k Arm Selection
PAC Subset Selection in Stochastic Multi-armed Bandits
lil' UCB : An Optimal Exploration Algorithm for Multi-Armed Bandits
Almost Optimal Exploration in Multi-Armed Bandits
On the Complexity of Best-Arm Identification in Multi-Armed Bandit Models
An optimal algorithm for the Thresholding Bandit Problem
Pure Exploration of Multi-armed Bandit Under Matroid Constraints
...
1
2
3
4
...