I am a Ph.D. student advised by Professor In So Kweon, at Electrical Engineering, KAIST.

My research interests lie in Machine Learning, Reinforcement Learning and Robotics.

My research topics include

  • Reinforcement Learning
  • Evolutionary Algorithms
  • Reinforcement Learning Combined with Computer Vision
  • Multi-agent Reinforcement Learning in Real-world Robotics System

Contact

  • kyunghyun.lee [at] kaist.ac.kr

  • Bldg N1, Rm 211, 291 Daehak-ro, Yuseong-gu, Daejeon, Korea, 34141

Eduction

  • MS in Aerospace Engineering, 2015

    Seoul National University (SNU), Korea

    Thesis: "Development of A* Real-time Path Planning Algorithm of UAV Considering Collision Avoidance

    Advisor: Youdan Kim

  • BS in Mechanical Engineering, 2012

    Pohang University of Science and Technology (Postech), Korea

    GPA: 3.81/4.3

Experiences

  • The Robotics Institute, Carnegie Mellon University, Pittsburgh PA, USA
    Jun 2019 - Aug 2019

    Visiting Researcher
    Advisor: Jean Oh, Martial Hebert
  • UVify, Inc., Seoul, Korea
    Mar 2015 - Jul 2018

    CPO
    Electoronics and software engineering in OORI and DRACO

Publications

  • Scalable Evolutionary Reinforcement Learning with Asynchronous Critic Update and Distributed Local Replay

    Kyunghyun Lee, UkcheolShin, Byeong-Uk Lee, In So Kweon

    Under review 2023

    [ paper / project / code ]

  • Autonomous Camera Exposure and Gain Control using Reinforcement Learning

    Kyunghyun Lee, UkcheolShin, Byeong-Uk Lee, In So Kweon

    Under review 2023

    [ paper / project / code ]

  • Complementary Random Masking for RGB-T Semantic Segmentation

    UkcheolShin, Kyunghyun Lee, In So Kweon

    Under review 2023

    [ paper / project / code ]

  • Self-supervised Monocular Depth Estimation from Thermal Images via Adversarial Multi-spectral Adaptation

    UkcheolShin, Kwanyong Park, Byeong-Uk Lee, Kyunghyun Lee, In So Kweon

    IEEE Winter Conference on Applications of Computer Vision (WACV) 2023, Best Student Paper Award

    [ paper / project / code / poster / award ]

  • Maximizing Self-supervision from Thermal Image for Effective Self-supervised Learning of Depth and Ego-motion

    UkcheolShin, Kyunghyun Lee, Byeong-Uk Lee, In So Kweon

    IEEE Robotics and Automation Letters (RA-L) 2022

    IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2022, Oral Presentation

    [ paper / project / code ]

  • DRL-ISP: Multi-Objective Camera ISP with Deep Reinforcement Learning

    UkcheolShin*, Kyunghyun Lee*, In So Kweon (* Equal contribution)

    IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2022, Oral Presentation

    [ paper / project ]

  • Self-supervised Depth and Ego-motion Estimation for Monocular Thermal Video using Multi-spectral Consistency Loss

    UkcheolShin, Kyunghyun Lee, Seokju Lee, In So Kweon

    IEEE Robotics and Automation Letters (RA-L) 2022

    IEEE International Conference on Robotics and Automation (ICRA) 2022, Oral Presentation

    [ paper / code / project ]

  • Depth Completion using Plane-Residual Representation

    Byeong-Uk Lee, Kyunghyun Lee, In So Kweon

    IEEE/CVF Conference on Computer Vision and Pattern Recognition 2021

    [ paper ]

  • An Efficient Asynchronous Method for Integrating Evolutionary and Gradient-based Policy Search

    Kyunghyun Lee, Byeong-Uk Lee, Ukcheol Shin, In So Kweon

    Conference on Neural Information Processing Systems 2020, Oral Presentation (Acceptance: 105/9454 ≈ 1.1%)

    [ paper / code ]

  • CNN-Based Simultaneous Dehazing and Depth Estimation,

    Byeong-Uk Lee, Kyunghyun Lee, Jean Oh, In So Kweon

    IEEE International Conference on Robotics and Automation 2020

    [ paper ]

  • Formation Flight of Multiple UAVs via Onboard Sensor Information Sharing

    Chulwoo Park, Namhoon Cho, Kyunghyun Lee, Youdan Kim

    Sensors 2015

    [ paper ]

Academic Activities

Reviewer

  • NeurIPS 2021-
  • ICLR 2022-
  • ICML 2022-
  • IROS 2022-
  • AAAI 2023-