Jongmin Lee
Assistant Professor in AI, Yonsei University
I am an assistant professor in AI at Yonsei University. Before joining Yonsei, I was a postdoctoral researcher at UC Berkeley, advised by Pieter Abbeel. I received my PhD from KAIST, where I was fortunate to be advised by Kee-Eung Kim. My research builds mathematically grounded yet practical reinforcement learning systems for reliable sequential decision making.
Education
2017 - 2022
Ph.D. in Computing, KAIST (Advisor: Kee-Eung Kim)
Thesis: Algorithms for Safe Reinforcement Learning
2015 - 2017
M.S. in Computing, KAIST (Advisor: Kee-Eung Kim)
Thesis: Constrained Bayesian Reinforcement Learning via Approximate Linear Programming
2009 - 2014
B.S. in Computer Science and Engineering, Seoul National University
Work Experience
Mar 2025 - Present
Assistant Professor in AI, Yonsei University
May 2022 - Feb 2025
Postdoctoral Researcher, UC Berkeley (Advisor: Pieter Abbeel)
Apr 2021 - Aug 2021
Research Scientist Intern, Google DeepMind (Host: Arthur Guez)
Awards & Honors
2022
Outstanding Ph.D. Thesis Award, School of Computing, KAIST
2020
Microsoft Research Asia Fellowship Nomination Award, Microsoft Research Asia
2019
Qualcomm-KAIST Innovation Awards (Paper Competition), Qualcomm
2018
Outstanding Presentation Award, 5th Society of Global Ph.D. Fellows Annual Conference
2018 - 2020
Global Ph.D. Fellowship, National Research Foundation of Korea
2017
Naver Ph.D. Fellowship, NAVER
Academic Service
NeurIPSArea Chair (2026), Reviewer (2016, 2018-2022, 2025)
ICMLArea Chair (2026), Reviewer (2019-2023)
ICLRArea Chair (2026), Reviewer (2020-2024)
CoRLArea Chair (2026)
AAAIProgram Committee (2020-2024, 2026, 2027)
COLMReviewer (2026)
ACL Rolling Review (ARR)Reviewer (2026)
IJCAIProgram Committee (2021-2022)
ACMLReviewer (2017, 2019, 2021)
Foundation Models for Decision Making Workshop @ NeurIPSReviewer (2022-2023)
Journal of Artificial Intelligence Research (JAIR)Reviewer (2019, 2025, 2026)
Machine Learning Journal (MLJ)Reviewer (2017, 2019)
Transactions on Machine Learning Research (TMLR)Reviewer (2023)
Operations Research LettersReviewer (2023)
Journal of Computational ScienceReviewer (2025)
Publications
2026
-
P3PROMISE: Proof Automation as Structural Imitation of Human ReasoningPreprint
-
C29ACPO: Agent-Chained Policy Optimization for Multi-Agent Reinforcement Learning
-
C28Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
2025
-
P2Group-Invariant Unsupervised Skill Discovery: Symmetry-aware Skill Representations for Generalizable BehaviorPreprint
-
P1Semi-gradient DICE for Offline Constrained Reinforcement LearningPreprint
-
C27FairDICE: Fairness-Driven Offline Multi-Objective Reinforcement Learning
-
C26SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation
2024
-
C23Mitigating Covariate Shift in Behavioral Cloning via Robust Stationary Distribution Correction
-
C22ROIDICE: Offline Return on Investment Maximization for Efficient Decision Making
-
C25Body Transformer: Leveraging Robot Embodiment for Policy Learning
-
C24Kernel Metric Learning for In-Sample Off-Policy Evaluation of Deterministic RL Policiespaper spotlight
2023
-
C21AlberDICE: Addressing Out-Of-Distribution Joint Actions in Offline Multi-Agent RL via Alternating Stationary Distribution Correction Estimation
-
C20SafeDICE: Offline Safe Imitation Learning with Non-Preferred Demonstrations
-
C19Tempo Adaptation in Non-stationary Reinforcement Learning
2022
-
C15LobsDICE: Offline Imitation Learning from Observation via Stationary Distribution Correction Estimation
-
C14Local Metric Learning for Off-Policy Evaluation in Contextual Bandits with Continuous Actions
-
C18COptiDICE: Offline Constrained Reinforcement Learning via Stationary Distribution Correction Estimation
-
C17DemoDICE: Offline Imitation Learning with Supplementary Imperfect Demonstrations
-
C16GPT-Critic: Offline Reinforcement Learning for End-to-End Task-Oriented Dialogue Systems
2021
-
C12,W5OptiDICE: Offline Policy Optimization via Stationary Distribution Correction Estimation
-
C11Representation Balancing Offline Model-based Reinforcement Learning
-
C13Monte-Carlo Planning and Learning with Language Action Value Estimates
2020
-
C7Reinforcement Learning for Control with Multiple Frequencies
-
C10Batch Reinforcement Learning with Hyperparameter Gradients
-
C8Monte-Carlo Tree Search in Continuous Action Spaces with Value Gradients
-
C9,W4Bayes-Adaptive Monte-Carlo Planning and Learning for Goal-Oriented Dialogues
2019
-
C5Trust Region Sequential Variational Inference
-
C6PyOpenDial: A Python-based Domain-Independent Toolkit for Developing Spoken Dialogue Systems with Probabilistic Rules
2018
-
C4Monte-Carlo Tree Search for Constrained POMDPs
-
W3Monte-Carlo Tree Search for Constrained MDPs
-
J1Layered Behavior Modeling via Combining Descriptive and Prescriptive Approaches: a Case Study of Infantry Company Engagement
2017
-
C3,W2Constrained Bayesian Reinforcement Learning via Approximate Linear Programming
-
C2Hierarchically-partitioned Gaussian Process Approximation
2016
-
W1Multi-View Automatic Lip-Reading using Neural Network
-
C1Bayesian Reinforcement Learning with Behavioral Feedback