A kernel-interpolation explore-then-commit algorithm achieves sublinear regret in high-dimensional contextual bandits under low-rank context covariance, and sublinear lenient regret under weaker spectral decay.
Thompson sampling for contextual bandits with linear payoffs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
stat.ML 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
High-dimensional Nonparametric Contextual Bandit Problem
A kernel-interpolation explore-then-commit algorithm achieves sublinear regret in high-dimensional contextual bandits under low-rank context covariance, and sublinear lenient regret under weaker spectral decay.