The paper establishes the first tilde O(epsilon^{-1}) upper bounds and matching lower bounds for forward-KL-regularized offline contextual bandits under single-policy concentrability in both tabular and general function approximation settings.
International Conference on Machine Learning , pages=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
New algorithms for joint contextual MNL assortment and pricing deliver improved online regret bounds of order W sqrt(d T log N)/L0 and local suboptimality guarantees offline.
DQPOPE estimates the entire return distribution in off-policy evaluation via deep quantile process regression, providing statistical advantages over standard single-value methods with equivalent sample sizes.
POOL is a new RL algorithm that adds privacy protection in continuous spaces with one-sided feedback and achieves sample complexity matching known non-private lower bounds.
RAFT aligns generative models by ranking samples with a reward model and fine-tuning only on the top-ranked outputs, reporting gains on reward scores and automated metrics for LLMs and diffusion models.
citing papers explorer
-
Fast Rates for Offline Contextual Bandits with Forward-KL Regularization under Single-Policy Concentrability
The paper establishes the first tilde O(epsilon^{-1}) upper bounds and matching lower bounds for forward-KL-regularized offline contextual bandits under single-policy concentrability in both tabular and general function approximation settings.
-
Optimal Online and Offline Algorithms for Contextual MNL with Applications to Assortment and Pricing
New algorithms for joint contextual MNL assortment and pricing deliver improved online regret bounds of order W sqrt(d T log N)/L0 and local suboptimality guarantees offline.
-
Distributional Off-Policy Evaluation with Deep Quantile Process Regression
DQPOPE estimates the entire return distribution in off-policy evaluation via deep quantile process regression, providing statistical advantages over standard single-value methods with equivalent sample sizes.
-
Privacy Preserving Reinforcement Learning with One-Sided Feedback
POOL is a new RL algorithm that adds privacy protection in continuous spaces with one-sided feedback and achieves sample complexity matching known non-private lower bounds.
-
RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment
RAFT aligns generative models by ranking samples with a reward model and fine-tuning only on the top-ranked outputs, reporting gains on reward scores and automated metrics for LLMs and diffusion models.