Joongkyu Lee
Joongkyu Lee
Home
Publications & Preprints
Light
Dark
Automatic
1
Nearly Minimax Optimal Regret for Multinomial Logistic Bandit
In this paper, we study the contextual multinomial logit (MNL) bandit problem in which a learning agent sequentially selects an …
Joongkyu Lee
,
Min-hwan Oh
PDF
Randomized Exploration for Reinforcement Learning with Multinomial Logistic Function Approximation
We study reinforcement learning with
multinomial logistic
(MNL) function approximation where the underlying transition probability …
Wooseong Cho
,
Taehyun Hwang
,
Joongkyu Lee
,
Min-hwan Oh
PDF
Demystifying Linear MDPs and Novel Dynamics Aggregation Framework
In this paper, we first challenge the common premise that linear MDPs always induce performance guarantees independent of the state …
Joongkyu Lee
,
Min-hwan Oh
PDF
Learning Uncertainty-Aware Temporally-Extended Actions
In reinforcement learning, temporal abstraction in the action space, exemplified by action repetition, is a technique to facilitate …
Joongkyu Lee
,
Seung Joon Park
,
Yunhao Tang
,
Min-hwan Oh
PDF
Cite
×