I am an AI research scientist at LG AI Research working with Prof. Honglak Lee. My current research focuses on the intersection of reinforcement learning and large language models.

I received my M.S./Ph.D. from MIT, studying multiagent reinforcement learning (doctoral thesis committee: Prof. Jonathan P. How, Prof. Jakob N. Foerster, and Prof. Pulkit Agrawal). Prior to my graduate study, I received a B.S. (summa cum laude) from Cornell and worked on computer vision research (advisor: Prof. Tsuhan Chen).

I am a recipient of Kwanjeong Educational Foundation Fellowship, and have spent time at CMU-RI (advisor: Prof. Sebastian Scherer) and TTIC (advisor: Prof. Matthew R. Walter).

🤗 My up-to-date CV is available upon request.

Selected News

• (Jan 2023) I successfully defended my doctoral thesis “Effective Learning in Non-Stationary Multiagent Environments” 🎉
• (Nov 2022) Our NeurIPS 2022 paper is featured on MIT News
• (Sep 2022) Our work on long-term MARL is accepted to NeurIPS 2022
• (Jun 2022) I presented our works on MARL at MILA’s MARL meeting
• (Jul 2021) I presented our works on meta-MARL at MILA’s RL sofa meeting
• (May 2021) Our work on meta-MARL is accepted to ICML 2021
• (Jan 2019) Our work on learning to teach is selected for outstanding student paper honorable mention at AAAI 2019 🎉

Publication

Preprint

• Anthony Z. Liu, Dong-Ki Kim, Sungryull Sohn, Honglak Lee. Learning Higher Order Skills that Efficiently Compose. In International Conference on Machine Learning (ICML) Workshop, 2023.

Reinforcement Learning

• Dong-Ki Kim, Matthew Riemer, Miao Liu, Jakob N. Foerster, Michael Everett, Chuangchuang Sun, Gerald Tesauro, Jonathan P. How. Influencing Long-Term Behavior in Multiagent Reinforcement Learning. In Neural Information Processing Systems (NeurIPS), 2022. [Paper] [Code] [Video] [MIT News]
  • Earlier version was presented at ICLR 2022 workshop with spotlight 🎉
• Chuangchuang Sun, Dong-Ki Kim, Jonathan P. How. ROMAX: Certifiably Robust Deep Multiagent Reinforcement Learning via Convex Relaxation. In International Conference on Robotics and Automation (ICRA), 2022. [Paper]
• Marwa Abdulhai, Dong-Ki Kim, Matthew Riemer, Miao Liu, Gerald Tesauro, Jonathan P. How. Context-Specific Representation Abstraction for Deep Option Learning. In Association for the Advancement of Artificial Intelligence (AAAI), 2022. [Paper] [Code] [Video]
• Dong-Ki Kim, Miao Liu, Matthew Riemer, Chuangchuang Sun, Marwa Abdulhai, Golnaz Habibi, Sebastian Lopez-Cot, Gerald Tesauro, Jonathan P. How. A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning. In International Conference on Machine Learning (ICML), 2021. [Paper] [Code] [Video]
• Chuangchuang Sun, Dong-Ki Kim, Jonathan P. How. FISAR: Forward Invariant Safe Reinforcement Learning with a Deep Neural Network-Based Optimizer. In International Conference on Robotics and Automation (ICRA), 2021. [Paper]
• Dong-Ki Kim, Miao Liu, Shayegan Omidshafiei, Sebastian Lopez-Cot, Matthew Riemer, Golnaz Habibi, Gerald Tesauro, Sami Mourad, Murray Campbell, Jonathan P. How. Learning Hierarchical Teaching in Cooperative Multiagent Reinforcement Learning. In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2020. [Paper] [WIRED News]
• Samir Wadhwania, Dong-Ki Kim, Shayegan Omidshafiei, Jonathan P. How. Policy Distillation and Value Matching in Multiagent Reinforcement Learning. In International Conference on Intelligent Robots and Systems (IROS), 2019. [Paper] [Video]
• Shayegan Omidshafiei, Dong-Ki Kim, Miao Liu, Gerald Tesauro, Matthew Riemer, Christopher Amato, Murray Campbell, Jonathan P. How. Learning to Teach in Cooperative Multiagent Reinforcement Learning. In Association for the Advancement of Artificial Intelligence (AAAI), 2019. [Paper] [MIT News]
  • Our work received outstanding student paper honorable mention at AAAI 2019 🎉
• Shayegan Omidshafiei*, Dong-Ki Kim*, Jason Pazis, Jonathan P. How. Crossmodal Attentive Skill Learner. In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2018 and Journal of Autonomous Agents and Multi-Agent Systems (JAAMAS), 2020. [Conference Paper] [Journal Paper] [Code] [Video]

Computer Vision & Robotics

• Lena M. Downes, Dong-Ki Kim, Ted J. Steiner, Jonathan P. How. City-wide Street-to-Satellite Image Geolocalization of a Mobile Ground Agent. In International Conference on Intelligent Robots and Systems (IROS), 2022. [Paper] [Video]
• Andrea Tagliabue, Dong-Ki Kim, Michael Everett, Jonathan P. How. Demonstration-Efficient Guided Policy Search via Imitation of Robust Tube MPC. In International Conference on Robotics and Automation (ICRA), 2022. [Paper] [Video]
• Dong-Ki Kim, Matthew R. Walter. Satellite Image-based Localization via Learned Embeddings. In International Conference on Robotics and Automation (ICRA), 2017. [Paper] [Video] [NVIDIA News]
• Dong-Ki Kim, Daniel Maturana, Masashi Uenoyama, Sebastian Scherer. Season-Invariant Semantic Segmentation with A Deep Multimodal Network. In Field and Service Robotics (FSR), 2017. [Paper]
• Hang Chu, Dong-Ki Kim, Tsuhan Chen. You Are Here: Mimicking the Human Thinking Process in Reading Floor-Plans. In International Conference on Computer Vision (ICCV), 2015. [Paper] [Video]
• Dong-Ki Kim, Tsuhan Chen. Deep Neural Network for Real-Time Autonomous Indoor Navigation. Technical Report, 2015. [Paper] [Video]