Regular biography
Beomjoon Kim is a faculty member in the Department of Computer Science at KAIST. Their research is associated with the Kim Jaechul Graduate School of AI. Kim's profile page indicates they have moved, and their contact information includes an email address and a website link. No detailed bio or additional information is provided.
Scholar-generated biography
Beomjoon Kim is a researcher at Korea Advanced Institute of Science & Technology (KAIST) specializing in Machine Learning, Robotics, and Artificial Intelligence. His work focuses on advancing robotic learning through datasets and models, with an emphasis on task and motion planning, imitation learning, and non-prehensile manipulation. Kim's research explores methods for guiding search in continuous state-action spaces, learning value functions with relational state representations, and developing frameworks for long-horizon planning. His contributions include the creation of Open X-Embodiment, a collaborative robotic learning platform, and the development of algorithms for socially adaptive path planning and Monte Carlo tree search in continuous spaces.