Publications

Top AI/CS Conference Proceedings

1. Guhyeon Kang, Minhae Kwon, “Variance-Averse n-Step Offline Reinforcement Learning for Sparse Long-Horizon Environments,” Advances in Neural Information Processing Systems (NeurIPS), December 2026. 

2. Guhyeon Kang, Jaehwi Lee, Minhae Kwon, “Simple Actors and Deep Critics for Scalable Reinforcement Learning,” ACM International Conference on Information and Knowledge Management (CIKM), November 2026.

Domestic Journal Articles

1. Guhyeon Kang, Jaehwi Lee, Eom Chanin, Minhae Kwon, “Offline Reinforcement Learning Based Autonomous Driving on Complex Roads with On-Ramp Merging and Bottleneck,” T-KSAE, vol.33, no.6, pp.465-479, 1, June 2025 

Domestic Conference Proceedings

1. Guhyeon Kang, Jaehwi Lee, Eom Chanin, Minhae Kwon, “Offline Reinforcement Learning Based Autonomous Driving Strategy for Complex Road Structures,” KSAE Spring, May, 2025.

2. Guhyeon Kang, Jaehwi Lee, Eom Chanin, Minhae Kwon, “Deep Reinforcement Learning Based Autonomous Driving Strategy for On-Ramp Merging and Bottleneck Traffic Scenarios,” KSAE Fall, Nov, 2024.

3. Guhyeon Kang, Jaehwi Lee, Eom Chanin, Minhae Kwon, “Autonomous Driving Strategy Using Offline Reinforcement Learning with Diverse Quality Driving Datasets,” KSAE Fall, Nov, 2024.

Machine Learning Workshop Papers

1. Sangeun Park, Guhyeon Kang, Minhae Kwon, "Improving Multi-Agent Coordination with a Drift-Aware RL Objective," ICML Workshop on DEMO, May 2025. (Poster)

Under Review & In Preparation

1. Guhyeon Kang*, Sangeun Park*, Chanin Eom, Jaehwi Lee, Minhae Kwon, "Offline Distributional Reinforcement Learning with Multi-step Returns for Combat Decision-making," in preparation.  (*equal contribution)

2. Guhyeon Kang, Sangeun Park, Minhae Kwon, “Anonymous submission – Flow-based Offline Reinforcement Learning,” IEEE International Conference on Robotics and Automation (ICRA), under review. 

3. Sangeun Park*, Guhyeon Kang*, Minhae Kwon, "Anonymous submission – Multi-agent LLM," Top AI Conference, under review. (*equal contribution)

Patent and Software Registration

Patent

1. Kang Guhyeon, Jaehwi Lee, Eom Chanin, Minhae Kwon, “Distributional reinforcement learning method for decision making in offline environment and computing device for performing same", Korean Patent, filed on Aug 28, 2025 (Application no. 10-2025-0121310)

Software Registration 

2. Kang Guhyeon, Minhae Kwon, “Offline Reinforcement Learning and Federated Post-Training for Personalization of Driving Characteristics in Autonomous Driving, August 12, 2025. (Registration no. C-2025-030854)

3. Kang Guhyeon, Minhae Kwon, “Autonomous vehicle training program for on-ramp merge and bottleneck strategy based on DDPG,” Korea Copyright Commission, October 16, 2024. (Registration no. C-2024-036832)

4. Kang Guhyeon, Minhae Kwon, “Visualization program for speeds by location of autonomous vehicles and non-autonomous vehicles at merges and bottlenecks,” Korea Copyright Commission, October 16, 2024. (Registration no. C-2024-036833)

5. Kang Guhyeon, Minhae Kwon, “Location-based autonomous driving policy training program,” Korea Copyright Commission, October 16, 2024. (Registration no. C-2024-036834)

6. Kang Guhyeon, Minhae Kwon, “Autonomous driving policy training program on merge and bottleneck roads with nonlinear traffic density coefficient,” Korea Copyright Commission, October 16, 2024. (Registration no. C-2024-036835)

7. Kang Guhyeon, Minhae Kwon,“A reinforcement learning-based autonomous driving policy learning program in an environment with human vehicles having different acceleration variations,” Korea Copyright Commission, October 16, 2024.