paper-with-me

홈 › Papers

Minimax Optimal Variance-Aware Regret Bounds for Multinomial Logistic MDPs

2026-05-19 · Pierre Boudart, Pierre Gaillard, Alessandro Rudi arxiv

We study reinforcement learning for episodic Markov Decision Processes (MDPs) whose transitions are modelled by a multinomial logistic (MNL) model. Existing algorithms for MNL mixture MDPs yield a regret of $\smash{\tilde{O}(dH^2\sqrt{T})}$ (Li et al., 2024), where $d$ is the feature dimension, $H$ the episode length, and $T$ the number of episodes. Inspired by the logistic bandit literature (Abeille et al., 2021; Faury et al., 2022; Boudart et al., 2026), we introduce a problem-dependent constant $\barσ\_T \leq 1/2$, measuring the normalised average variance of the optimal downstream value function along the learner's trajectory. We propose an algorithm achieving a regret of $\smash{\tilde{O}(dH^2\barσ\_T\sqrt{T})}$, which recovers the existing bound in the worst case and improves upon it for structured MDPs. For instance, for KL-constrained robust MDPs, $\barσ\_T = O(H^{-1})$, reducing the horizon dependence by a factor $H$. We further establish a matching $\smash{Ω(dH^2\barσ\_T\sqrt{T})}$ lower bound, proving minimax optimality (up to logarithmic factors) and fully characterising the regret complexity of MNL mixture MDPs for the first time.

📄 PDF Abstract BibTeX arXiv:2605.19768

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Minimax Regret for Cascading Bandits

2022-03-23 · Daniel Vial, Sujay Sanghavi, Sanjay Shakkottai, R. Srikant

Cascading bandits is a natural and popular model that frames the task of learning to rank from Bernoulli click feedback in a bandit setting. For the case of unstructured rewards, we prove matching upper and lower bounds …

Learning-To-Rank

Sharp Variance-Dependent Bounds in Reinforcement Learning: Best of Both Worlds in Stochastic and Deterministic Environments

2023-01-31 · Runlong Zhou, Zihan Zhang, Simon S. Du

We study variance-dependent regret bounds for Markov decision processes (MDPs). Algorithms with variance-dependent regret guarantees can automatically exploit environments with low variance (e.g., enjoying constant regre…

Variance-Aware Sparse Linear Bandits

2022-05-26 · Yan Dai, Ruosong Wang, Simon S. Du

It is well-known that for sparse linear bandits, when ignoring the dependency on sparsity which is much smaller than the ambient dimension, the worst-case minimax regret is $\widetilde{\Theta}\left(\sqrt{dT}\right)$ wher…

Asymptotically Optimal Fixed-Budget Best Arm Identification with Variance-Dependent Bounds

2023-02-06 · Masahiro Kato, Masaaki Imaizumi, Takuya Ishihara, Toru Kitagawa

We investigate the problem of fixed-budget best arm identification (BAI) for minimizing expected simple regret. In an adaptive experiment, a decision maker draws one of multiple treatment arms based on past observations …

Towards Optimal Differentially Private Regret Bounds in Linear MDPs

2025-04-12 · Sharan Sahu

We study regret minimization under privacy constraints in episodic inhomogeneous linear Markov Decision Processes (MDPs), motivated by the growing use of reinforcement learning (RL) in personalized decision-making system…

Offline RLReinforcement Learning (RL)