Welcome, everybody!
I am a graduate student at Brain and Machine Intelligence Lab (BMIL) (Prof. Minhae Kwon), Soongsil Graduate School in Seoul, South Korea. My present research topic is a reinforcement learning algorithm development.
I am currently interested in all fields of learning-based control (e.g., Online RL, Offline RL, Imitation Learning, Inverse RL, and so on), Cognitive science (or psychology), and Neuroscience. In detail, the decision-making system of autonomous mobile devices (vehicles, drones, etc.).
Previously, I majored in Biotechnology and Bioinformatics at the School of Systems Biomedical Science at Soongsil Univ in Seoul, South Korea. Additionally, I had a double major in Big Data Computing Systems.
I interned at Post-Genome Informatics Lab (Prof. Sangsoo Kim) at the School of Systems Biomedical Science, Soongsil University.
Be able to confirm my CV if you want.
Publication (J: Journal; C: Conference; W: Workshop; T: Thesis)
[J5] Dongsu Lee, Minhae Kwon, “Instant Inverse Modeling of Stochastic Driving Behavior with Deep Reinforcement Learning,” IEEE Transactions on Consumer Electronics, accepted.
[J4] Chanin Eom, Dongsu Lee, Minhae Kwon, “Selective Imitation for Efficient Online Reinforcement Learning with Pre-collected Data,” ICT Express, vol. 11, no. 2, 2025.
[J3] Jeyeon Eo, Dongsu Lee, and Minhae Kwon, “The Impact of Dataset on Offline Reinforcement Learning Performance in UAV-based Emergency Network Recovery Tasks,” IEEE Communications Letters, vol. 28, no. 5, May 2024.
[J2, W1] Dongsu Lee, Minhae Kwon, “ Stability Analysis in Mixed-autonomous Traffic with Deep Reinforcement Learning,” IEEE Transactions on Vehicular Technology, vol 72, no. 3, pp. 2848 – 2862, March 2023. (IF 6.239, Rank 38/276, Top 13.59%) [Media] NeurIPS DeepRL Workshop [Media] [Video]
[J1] Dongsu Lee, Minhae Kwon, “ ADAS-RL: Safety Learning Approach for Stable Autonomous Driving,” ICT Express, vol.8, no.3, September 2022. (IF 4.317, Rank 22/91, Top 23.26%)
[C3] Dongsu Lee, Minhae Kwon, “Episodic Future Thinking Mechanism for Multi-agent Reinforcement Learning” Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, December 2024 (Acceptance rate 25.8%) project page|pre-print|talk (5min).
[C2] Dongsu Lee, Chanin Eom, Minhae Kwon, “AD4RL: Autonomous Driving Benchmarks for Offline Reinforcement Learning with Value-based Dataset” IEEE International Conference on Robotics and Automation (ICRA), Yokohama, Japan, May 2024. project page|pre-print|talk (5min)
[C1] Dongsu Lee, Minhae Kwon, “Foresighted Decisions for Inter-vehicle Interactions: A Offline Reinforcement Learning Approach” IEEE Intelligent Transportation System Conference (IEEE ITSC), Bilbao, Spain, September 2023. project papge|talk (10min)
[T-BS] Dongsu Lee, Chayoung Lee, Expression Quantitative Trait Loci for CCR4-NOT Complex that Regulate Global Gene Expression, Soongsil University, 2022. [pdf]
Under Review
Dongsu Lee, Minhae Kwon, “Offline Reinforcement Learning on Foresighted Decision-making for Autonomous Driving System,” under review.
Chanin Eom, Dongsu Lee, Minhae Kwon, “Price of the Autonomous Strategy with Reinforcement Learning in Mixed-autonomous Traffic,” under review.
Dongsu Lee, Minhae Kwon, “Scenario-free Autonomous Driving with Multi-task Offline-to-online Reinforcement Learning,” under review.
Dongsu Lee, Minhae Kwon, Anonymous submission - multi-task reinforcement learning, Top AI Conference, under review.
Dongsu Lee, Minhae Kwon, Anonymous submission – offline reinforcement learning, Top AI Conference, under review.
Issued Patents
Dongsu Lee, Minhae Kwon, “Method for Combating Stop-and-Go Wave Problem Using Deep Reinforcement Learning based Autonomous Vehicles, Recording medium and device for performing this the method,” US Patent, Nov 27, 2021. (Application no. US 17/535567), issued on September 17, 2024. (Patent no. US 12,091,025)
숭실대학교 산학협력단 “자율주행차량을 위한 심층강화학습 기반 차선 변경 방법, 이를 수행하기 위한 기록 매체 및 장치,” Korean Patent, filed on Sep 06, 2021 (Application no. 10-2021-0118540), issued on January 25, 2024. (Patent no. 10-2631402) Developer: Minhae Kwon, Dongsu Lee [PDF]
숭실대학교 산학협력단, “심층강화학습기반 자율주행차량을 이용한 정체 현상 해결 방법, 이를 수행하기 위한 기록 매체 및 장치,” Korean Patent, filed on July 13, 2021 (Application no. 10-2021-0091665), issued on October 19, 2022. (Patent no. 10-2457914) Developer: Minhae Kwon, Dongsu Lee [PDF]
Travel and Grant
I am sponsored by IITP & Sogang Univ. for AI Intensive Program at Carnegie Mellon University (USD 41K; 2024/08~2025/02), and you can check our daily in Pittsburgh, PA, USA, through 5744_woodmont.
I am receiving Future Industrial Talent scholarship support from CMK Hyundai Group. (2023/08/15~)
I am a group member of 18th Qualcomm IT Tour, and you can check our travels in San Diego, California, USA (2022/08/15~21), through Instagram-Team b.
I was invited two times to the 2022/2024 AI Conference hosted by Alberta Machine Intelligence Institute (AMII), Edmonton, Canada (2022/05/22~29 and 2024/05/20~25).
I have completed 2021 Huawei Seeds for the Future [media], Virtual.
News
2024.10. Won NeurIPS 2024 Scholar Award (Full conference registration and accommodations).
2024.09. Accepted conference paper in Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada.
2024.09. Accepted our journal to IEEE Transactions on Consumer Electronics.
2024.08. Accepted our letter to ICT Express.
2024.05. Accepted five workshop papers in IEEE ICRA, Yokohama, Japan.
2024.05. Won the Haedong Best Paper Award at KICS 2024 Summer Conference, South Korea.
2024.04. Won 2024 ICRA Travel Grant (USD 0.5K).
2024.03. Won AI Intensive Program at Carnegie Mellon University sponsored by IITP & Sogang Univ. (USD 41K).
2024.02. Issued Korean patent Dongsu Lee, Minhae Kwon, “METHOD OF LANE CHANGE FOR AUTONOMOUS VEHICLES BASED DEEP REINFORCEMENT LEARNING, RECORDING MEDIUM AND DEVICE FOR PERFORMING THE METHOD”
2024.01. Accepted conference paper in IEEE ICRA, Yokohama, Japan.
2024.01. Our domestic journal paper, Online Curriculum Reinforcement Learning Based UAVTrainingfor Disaster Network Recover, is selected as Best paper in J-KICS [Web]
2024.01. Won the Best Paper Award at KICS 2023 Fall Conference, South Korea [interview]
2023.12. Invited by 2024 AI Conference hosted by AMII, Edmonton, Canada.
2023.11. Accepted our letter to IEEE Communications Letters (CL).
2023.09. Won the President’s Award of IITP (Institute for Information & communication Technology Planning & evaluation) at ICT Challenge 2023. Dongsu Lee, Chanin Eom, Jaehwi Lee [media]
2023.08. Completed Research Grants (joined as Practice Responsibility) with Hyundai Motor Company (Hyundai NGV) and Accepted domestic Journal (J-KICS) “Survey on Practical Reinforcement Learning: from Imitation Learning to Offline Reinforcement Learning,” Dongsu Lee, Chanin Eom, Sungwoo Choi, Sungkwan Kim, and Minhae Kwon.
2023.07. Our domestic journal paper, Autonomous Driving Strategy for Bottleneck Traffic with Prioritized Experience Replay, is selected as Best paper in J-KICS [Web]
2023.07. Accepted conference papers in 2023 IEEE ITSC
2023.06. Presented at EIRIC Seminar [Web]
2023.06. Accepted four papers in 2023 ICML Workshop
2023.06. Got a 2023 Hyundai On-dream Future Industrial Talent Scholarship
2023.05. Presented our seven patent at Industrial Technology R&D Symposium [youtube]
2023.01. Won the Haedong Best Paper Award at KICS 2023 Winter Conference, South Korea
2022.10. Issued Korean patent Dongsu Lee, Minhae Kwon, “Method for Combating Stop-and-Go Wave Problem Using Deep Reinforcement Learning based Autonomous Vehicles, Recording medium and device for performing this the method” [PDF]
2022.10. Accepted Dongsu Lee, Minhae Kwon, “Stability Analysis in Mixed-autonomous Traffic with Deep Reinforcement Learning” to IEEE Transactions on Vehicular Technology (TVT)
2022.07. Joined in Qualcomm IT Tour as 18th group member. See you in San Diego, California, USA in August 2022
2022.05. Accepted Dongsu Lee, Minhae Kwon, “ADAS-RL: Safety Learning Approach for Stable Autonomous Driving” to ICT express
2022.04. Invited by 2022 AI Week hosted by AMII, Edmonton, Canada
2022.03. Graduate Admission, School of Electronic Engineering, Soongsil Graduate School, Seoul, South Korea
2022.02. Won the Best Paper Award at KICS 2022 Winter Conference, South Korea
2022.02. Graduated researcher, Brain and Machine Intelligence Lab, School of Electronic Engineering, Soongsil University, Seoul, South Korea
2021.11. Won the Soongsil Patent Competition
2021.10. Accepted Dongsu Lee, Minhae Kwon, “Stability Analysis in Mixed-Autonomous Traffic with Deep Reinforcement Learning, to NeurIPS DeepRL Workshop
2021.10. Published Dongsu Lee, Minhae Kwon, “Combating Stop-and-Go Wave Problem at a Ring Road Using Deep Reinforcement Learning Based Autonomous Vehicles” to JKICS
2021.09. Joined Huawei Seeds for the Future 2021 (ICT Talent Nurturing Program) as a member
2021.06. Won the ICT Service Idea Competition hosted by KICS with Hyoseon Kye and Miru Kim
2020.12. Undergraduate Intern, Brain and Machine Intelligence Lab, School of Electronic Engineering, Soongsil University, Seoul, South Korea
2019.01. Undergraduate Intern, Post-Genome Informatics Lab, School of Systems Biomedical Science, Soongsil University, Seoul, South Korea
2016.03. Undergraduate Admission, School of Systems Biomedical Science, Soongsil University, Seoul, South Korea