I am currently at AI Institute Seoul National University (AIIS), completing my third year of service as Professional Research Personnel. I received my Ph.D. in the Interdisciplinary Program in Artificial Intelligence at Seoul National University in August 2026. During my graduate studies, I conducted research at the Biointelligence Lab under the supervision of Prof. Byoung-Tak Zhang, beginning in 2020.

Research Focus

My research focuses on Reinforcement Learning (RL) for sequential decision-making, with an emphasis on developing general learning methods that support effective behavior in autonomous systems. My work spans continuous control, visually guided tasks, and multi-agent tactical coordination.

• 🧠 Sequential Decision-Making & Reinforcement Learning
• 🛩️ Autonomous Systems & Multi-Agent Coordination
• 📷 Visual Learning & Control

I am increasingly interested in connecting advances in RL to practical autonomous systems, including autonomous combat systems and manned–unmanned teaming.

My doctoral research focused on a generalized framework for desired-state-based reinforcement learning to broaden the scope of application and the practical utility of RL.

Publications

• Exponential Increase in Rewards Boosts Reinforcement Learning
  Ganghun Lee, Minji Kim, Minsu Lee, Byoung-Tak Zhang
  (Under review)

• Climb with SHERPA: Heuristic-Guided Reinforcement Learning via Segmented Experience Relay
  Minji Kim, Ganghun Lee, Minsu Lee, Byoung-Tak Zhang
  RA-L 2026

• Recency-Biased Sampling for Up-to-Date Adaptation of Reinforcement Learning
  Ganghun Lee, Minji Kim, Minsu Lee, Byoung-Tak Zhang
  KSC 2025 (Best paper award) [pdf]

• Truncated Gaussian Policy for Debiased Continuous Control
  Ganghun Lee, Minji Kim, Minsu Lee, Byoung-Tak Zhang
  AAAI 2025 (Oral presentation) [pdf]

• Heuristic Action Cycle Reinforcement Learning
  Minji Kim, Ganghun Lee, Minsu Lee, Byoung-Tak Zhang
  KSC 2025 [pdf]

• Truncated Gaussian Policy Gradient for Bounded Continuous Action Space
  Ganghun Lee, Minji Kim, Byoung-Tak Zhang
  KCC 2024 [pdf]

• Bottleneck-Aware Linear Augmentation for Robotic Imitation Learning
  Minji Kim, Ganghun Lee, Hyunseo Kim, Minsu Lee, Byoung-Tak Zhang
  KCC 2024 [pdf]

• Neural Collage Transfer: Artistic Reconstruction via Material Manipulation
  Ganghun Lee, Minji Kim, Yunsu Lee, Minsu Lee, Byoung-Tak Zhang
  ICCV 2023 [pdf] [github]

• Defiant Policy Learning for Efficient Exploration in Reinforcement Learning
  Ganghun Lee, Minsu Lee, Byoung-Tak Zhang
  KSC 2023 [pdf]

• H2RL: Transferring Heuristic Knowledge to a Reinforcement Learning Agent for Solving the Online 3D Bin Packing Problem
  Minji Kim^†, Ganghun Lee^†, Minsu Lee, Byoung-Tak Zhang
  ICRA 2023 Workshop (Spotlight) [pdf]

• Multi-Agent Reinforcement Learning for Cover Tactics in SMAC Environment
  Yesol Park, Ganghun Lee, Yunsu Lee, Junseok Park, Byoung-Tak Zhang
  KSC 2023 [pdf]

• From Scratch to Sketch: Deep Decoupled Hierarchical Reinforcement Learning for Robotic Sketching Agent
  Ganghun Lee, Minji Kim, Minsu Lee, Byoung-Tak Zhang
  ICRA 2022 [pdf]

• Reinforcement Learning with Heuristics for Buffer-Utilized Random Bin Packing Problem
  Minji Kim, Ganghun Lee, Byoung-Tak Zhang
  Journal of Logistics Science & Technology (JLST) 2022 (Best paper award) [pdf]

• Deep Reinforcement Learning for Stroke-Conditioned Auto-Painting
  Ganghun Lee, Minsu Lee, Byoung-Tak Zhang
  KCC 2022 [pdf]

• Reset-free Multi-agent Reinforcement Learning Using Reversibility Estimation
  Minji Kim, Ganghun Lee, Byoung-Tak Zhang
  KCC 2022 [pdf]

• Toddler-Guidance Learning: Impacts of Critical Period on Multimodal AI Agents
  Junseok Park, Kwanyoung Park, Hyunseok Oh, Ganghun Lee, Minsu Lee, Youngki Lee, Byoung-Tak Zhang
  ICMI 2021 [pdf]

• Response Analysis of Deep Neural Networks to Optical Illusion Video
  Ganghun Lee, Kibeom Kim, Minsu Lee, Byoung-Tak Zhang
  KSC 2021 [pdf]

• Diverged Episodic Memory-based Self-Supervised Policy Learning
  Ganghun Lee, Kibeom Kim, Dong-sig Han, Byoung-Tak Zhang
  KCC 2021, KIISE 2023 (Best paper award) [pdf]

• Model-Based Curriculum Reinforcement Learning for Continuous Line Drawing
  Ganghun Lee, Minji Kim, Byoung-Tak Zhang
  KCC 2021 [pdf]

• Deep Reinforcement Learning for Image-Based Target Trajectory Decomposition
  Ganghun Lee, Byoung Tak Zhang
  The Korea Institute of Military Science and Technology (KIMST 2021)

• Reset-free Competitive reinforcement learning to learn robotic manipulation skills
  Minji Kim, Ganghun Lee, Kibeom Kim, Minsu Lee, Byoung-Tak Zhang
  KCC 2021 [pdf]

• Reward-based Contrastive Learning for Generalization in Deep Reinforcement Learning
  Inwoo Hwang, HoJoon Song, Dong-sig Han, Ganghun Lee, Byoung-Tak Zhang
  KCC 2021 [pdf]

• Evolved Intrinsic Motivation for Fully Autonomous Reinforce Learning
  Ganghun Lee, Byoung-Tak Zhang
  KSC 2020 [pdf]

• Multi-robot Map Merging with Pose Estimation and Image Generation based on Neural Network
  Jaein Kim, Hyundo Lee, Ganghun Lee, Yoonsung Kim, Dong-sig Han, Byoung-Tak Zhang
  KSC 2020 [pdf]

• Query-Based Zero-Shot Detection
  HoJoon Song^†, Ganghun Lee^†, Byoung-Tak Zhang
  KCC 2020 [pdf]

• Resolving Data Interoperability in Ubiquitous Health Profile Using Semi-Structured Storage and Processing
  Fahad Ahmed Satti, Wajahat Ali Khan, Ganghun Lee, Asad Masood Khattak, Sungyoung Lee
  ACM/SIGAPP SAC 2019 [pdf]

• Student to Teacher: Repeated Model Compression Through Reproducing Over the Shoulder Learning
  Ganghun Lee, Sungho Bae
  KSC 2019 Undergraduate Session (Best paper award) [pdf]

• A New YouTube View Count Prediction Method using DeepAudio-Video Multimodal Learning
  Haeyeon Park, Ganghun Lee, Yeonwoo Jang, Hyeongseok Kim, Sungho Bae
  KCC 2019 Undergraduate Session [pdf]

• Design and Implementation of OTP-based Digital Door Lock System
  Ganghun Lee, Seongbeom Hong, Jinsung Cho
  KSC 2017 Undergraduate Session [pdf]

Research & Industry Experience

military_tech Professional Research Personnel, Republic of Korea ^{2024.09. - 2027.08. (expected)}

- AI Institute Seoul National University (AIIS) ^{2026.09. - Present}

- Biointelligence Lab, Seoul National University ^{2024.09. - 2026.08.}

gps_fixed National Defense R&D Project with Hyundai Rotem ^{2023 - Present}

- Co-Lead Student Researcher

- RL-based multi-agent tactical algorithm development for tanks

flight National Defense R&D Project with Korea Aerospace Industries (KAI) ^{2024 - 2025}

- Lead Student Researcher

- RL-based multi-agent mission execution and tactical engagement for aircraft

science Government-Funded Project (NRF) ^{2021 - 2025}

- Student Researcher

- Participated in RL-focused research initiatives

precision_manufacturing Early Team Member at Tommoro Robotics ^{2022 - 2023}

- Robot AI Scientist

- Sim2Real logistics project using robotic arms

schema Research Intern at Surromind ^{2021 - 2023}

- AI Engineer

- MLOps platform development

local_laundry_service Industry–Academia Collaborative Research Project with LG Electronics ^{2020 - 2021}

- Student Researcher

- Laundry appliance user behavior analysis

scatter_plot National Defense R&D Project (BMRR) ^{2020 - 2021}

- Student Researcher

- 6D pose estimation model development

child_care Government-Funded Project (BabyMind) ^{2020 - 2021}

- Student Researcher

- NeurIPS 2020 workshop organization

Awards & Scholarships

• Grand Prize at 2025 Demo Competition, Seoul National University, 2025
• AAAI 2025 Oral Presentation, 2025
• KSC 2025 Best Paper Award, 2025
• ICRA 2023 Workshop Spotlight Presentation, 2023
• Best Paper Award, Journal of Logistics Science & Technology, 2022
• KCC 2021 Best Paper Award, 2021
• Academic Excellence Scholarship, Seoul National University, 2020, 2021
• KSC 2019 Undergraduate Paper Award (Encouragement Award), 2019
• KTB Startup Competition (Encouragement Award), 2019
• TOPCIT Excellence Scholarship, Kyunghee University, 2019
• Leadership & Activity Scholarship, Kyunghee University, 2019
• Programming Contest (Encouragement Award), Kyunghee University, 2018
• Academic Excellence Scholarship, Kyunghee University, 2016, 2017, 2019