Sunghoon Hong

Hi, I’m Sunghoon Hong, an Applied Research Scientist at LG AI Research.

I specialize in Deep Reinforcement Learning (DRL) and Multi-Agent Reinforcement Learning (MARL). My research focuses on Physical AI, developing advanced AI models to solve complex optimization and control problems in real-world industrial environments, such as manufacturing plant and robotics.

career

Research Scientist — LG AI Research

Seoul, Korea • 2022 – Present

Conducting research on physical AI
Developed reinforcement learning methods for petrochemical plant scheduling optimization (recognized with the 2025 LG Award)
Designed combinatorial optimization approaches for circuit design

news

Aug 16, 2025	Presented at IJCAI 2025 in Montreal, Canada
Apr 09, 2025	Received the 2025 LG Awards for petrochemical plant scheduling optimization
May 05, 2024	Presented at AAMAS 2024 in Auckland, New Zealand

latest posts

Oct 07, 2025	Hello, World!

selected publications

ICLR

Structure-Aware Transformer Policy for Inhomogeneous Multi-Task Reinforcement Learning

Sunghoon Hong, Deunsol Yoon, and Kee-Eung Kim

In International Conference on Learning Representations (ICLR), 2022

Bib HTML PDF Code

@inproceedings{hong2022structure,
  title = {Structure-Aware Transformer Policy for Inhomogeneous Multi-Task Reinforcement Learning},
  author = {Hong, Sunghoon and Yoon, Deunsol and Kim, Kee-Eung},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2022},
}

ICLR

Winning the L2RPN Challenge: Power Grid Management via Semi-Markov Afterstate Actor-Critic

Deunsol Yoon^*, Sunghoon Hong^*, Byung-Jun Lee, and Kee-Eung Kim

In International Conference on Learning Representations (ICLR), 2021

Spotlight (5%)

Bib HTML PDF

@inproceedings{yoon2021winning,
  title = {Winning the L2RPN Challenge: Power Grid Management via Semi-Markov Afterstate Actor-Critic},
  author = {Yoon, Deunsol and Hong, Sunghoon and Lee, Byung-Jun and Kim, Kee-Eung},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2021},
  note = {Spotlight (5%)},
}

ICML

Agent-Centric Actor-Critic for Asynchronous Multi-Agent Reinforcement Learning

Whiyoung Jung^*, Sunghoon Hong^*, Deunsol Yoon^*, Kanghoon Lee, and 1 more author

In International Conference on Machine Learning (ICML), 2025

Bib HTML PDF

@inproceedings{jung2025agent,
  title = {Agent-Centric Actor-Critic for Asynchronous Multi-Agent Reinforcement Learning},
  author = {Jung, Whiyoung and Hong, Sunghoon and Yoon, Deunsol and Lee, Kanghoon and Lim, Woohyung},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
}

ICML

Penalizing Infeasible Actions and Reward Scaling in Reinforcement Learning with Offline Data

Jeonghye Kim, Yongjae Shin, Whiyoung Jung, Sunghoon Hong, and 4 more authors

In International Conference on Machine Learning (ICML), 2025

Spotlight (2.6%)

Bib HTML PDF

@inproceedings{kim2025penalizing,
  title = {Penalizing Infeasible Actions and Reward Scaling in Reinforcement Learning with Offline Data},
  author = {Kim, Jeonghye and Shin, Yongjae and Jung, Whiyoung and Hong, Sunghoon and Yoon, Deunsol and Sung, Youngchul and Lee, Kanghoon and Lim, Woohyung},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  note = {Spotlight (2.6%)},
}

AAMAS

Naphtha Cracking Center Scheduling Optimization using Multi-Agent Reinforcement Learning

Sunghoon Hong, Deunsol Yoon, Whiyoung Jung, Jinsang Lee, and 8 more authors

In Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2024

Bib HTML PDF

@inproceedings{hong2024naphtha,
  title = {Naphtha Cracking Center Scheduling Optimization using Multi-Agent Reinforcement Learning},
  author = {Hong, Sunghoon and Yoon, Deunsol and Jung, Whiyoung and Lee, Jinsang and Yoo, Hyundam and Ham, Jiwon and Jung, Suhyun and Moon, Chanwoo and Jung, Yeontae Jung and Lee, Kanghoon and Lim, Woohyung and others},
  booktitle = {Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS)},
  year = {2024},
}