Jung-hun Kim (김정훈) [CV]
Postdoctoral Researcher at CREST, ENSAE, IP Paris – FairPlay team
junghunkim7786@gmail.com | junghun.kim@ensae.fr
[LinkedIn], [Google Scholar]

Research Overview 
I work on online learning and decision-making under uncertainty and non-stationarity, with applications to optimal stopping, matching, queueing systems, dynamic pricing, and recommendation.

My research combines theoretical guarantees with practically motivated models to study dynamic decision-making in complex systems. 

Academic Positions

• Postdoctoral Researcher, CREST, ENSAE, IP Paris – FairPlay team, France
  June 2025 - Present
  Supervisor: Vianney Perchet
  Research focus: Optimal stopping and prophet inequalities with learning and non-stationary rewards

• Postdoctoral Researcher, IDIS, Seoul National University, South Korea
  Sep 2023 - May 2025
  Supervisor: Min-hwan Oh
  Research focus: Online learning for matching and dynamic pricing under preference feedback

• Ph.D. in Industrial and Systems Engineering, KAIST,  South Korea
  Sep 2018 - Aug 2023
  Advisor: Se-Young Yun
  Research focus: Online learning under non-stationary environments

• Research intern, Department of Statistics, London School of Economics, UK
  Mar 2021 - Jul 2021
  Supervisor: Milan Vojnović
  Research focus: Online learning for scheduling in dynamic systems

Publications

• [ICLR 2026, to appear] Learning in Prophet Inequalities with Noisy Observations

Jung-hun Kim, Vianney Perchet

• [NeurIPS 2025] Oracle-Efficient Combinatorial Semi-Bandits

Jung-hun Kim, Milan Vojnović, Min-hwan Oh

• [ICML 2025] Tracking Most Significant Shifts in Infinite-Armed Bandits

Joe Suk, Jung-hun Kim

• [ICLR 2025] Dynamic Assortment Selection and Pricing with Censored Preference Feedback

Jung-hun Kim, Min-hwan Oh

• [TMLR 2025] Adversarial Bandits Against Arbitrary Strategies

Jung-hun Kim, Se-Young Yun

• [NeurIPS 2024] Queueing Matching Bandits with Preference Feedback

    Jung-hun Kim, Min-hwan Oh

• [NeurIPS 2024] An Adaptive Approach for Infinitely Many-armed Bandits under Generalized Rotting Constraints

Jung-hun Kim, Milan Vojnović, Se-Young Yun

• [AISTATS 2023] Contextual Linear Bandits under Noisy Features: Towards Bayesian Oracles

Jung-hun Kim, Se-Young Yun, Minchan Jeong, Junhyun Nam, Jinwoo Shin, Richard Combes

• [ICML 2022] Rotting Infinitely Many-armed Bandits

Jung-hun Kim, Milan Vojnović, Se-Young Yun

• [IEEE BigData 2018] Research Hypothesis Generation using Link Prediction in a Bipartite Graph

Jung-hun Kim, Aviv Segev

Working Papers 

• Matching Optimization with Preference Feedback (link)
  Jung-hun Kim, Min-hwan Oh

• Scheduling with Bilinear Rewards (link)
  Jung-hun Kim, Milan Vojnović

Academic Activities

• • Reviewer: NeurIPS, ICML, ICLR, AISTATS, JMLR, CDC, AAAI