Hyunwoo Ryu
About
I am an incoming EECS Ph.D. student at MIT CSAIL advised by Prof. Vincent Sitzmann. Previously, I obtained my MS degree in Artificial Intelligence and BA degree in Mechanical Engineering from Yonsei University, advised by Prof. Jongeun Choi. My research interests lie in incorporating spatio-temporal understandings to robot learning methods. Especially, I explore the mathematical tools from theoretical physics and differential geometry to gain deeper insights in modeling the spatio-temporal structures of robotic tasks. For example, I use Lie group and representation theoretic approaches to exploit the spatial roto-translational symmetry, also known as the SE(3)-equivariance, in robotic manipulation learning. My overarching aim is to conceive mathematical models that can accurately represent and model the causal structure of robot learning problems that lies within the space-time. This exploration resonates with my conviction that the robot itself is an embodiment within this space-time.
Publications
Conference Papers
Hyunwoo Ryu, Jiwoo Kim, Hyunseok An, Junwoo Chang, Joohwan Seo, Taehan Kim, Yubin Kim, Chaewon Hwang, Jongeun Choi, Roberto Horowitz
CVPR 2024
Highlight (11.9% of accepted papers)
Paper / Project Page / Code / Video
Hyunwoo Ryu, Hong-in Lee, Jeong-Hoon Lee, Jongeun Choi
ICLR 2023
Paper / Project Page / Codes / Video / Poster
Workshop Papers
(* indicates equal contribution)
Junwoo Chang*, Hyunwoo Ryu* (Equal Contribution), Jiwoo Kim, Soochul Yoo, Joohwan Seo, Nikhil Potu Surya Prakash, Jongeun Choi, Roberto Horowitz
NeurIPS 2023 Workshop on Diffusion Models
Jiwoo Kim*, Hyunwoo Ryu* (Equal Contribution, Oral Speaker), Jongeun Choi, Joohwan Seo, Nikhil Potu Surya Prakash, Ruolin Li, Roberto Horowitz
RSS 2023 Workshop on Symmetries in Robot Learning
Oral / Best Paper Award
Projects
Team Leader
The 5th Yonsei University Mechanical Engineering Graduate Student Academic Conference, 2023
Best Technical Demonstration Award
System Architect / ROS, SLAM, Navigation, Motion Planning, Sim2real Engineer
ICRA 2019 RoboMaster AI Challenge @Montreal, Canada
3rd Prize, S+ rank in Technical Reports (top 4 out of 68 universities worldwide)
News
[Apr. 2024] I'll start my Ph.D. in MIT EECS, beginning Fall 2024.
[Apr. 2024] Our paper "Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation" has been selected as a Highlight (11.9% of accepted papers) at CVPR 2024.
[Feb. 2024] Our paper "Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation" has been accepted to CVPR 2024.
[Oct. 2023] Our paper "Denoising Heat-inspired Diffusion with Insulators for Collision Free Motion Planning" has been accepted to NeurIPS 2023 Workshop on Diffusion Models.
[Oct. 2023] We won the Best Technical Demonstration Award for Diffusion-EDFs at the 5th Yonsei University Mechanical Engineering Graduate Student Academic Conference.
[Sept. 2023] Our new paper "Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation" available on Arxiv.
[July 2023] We won the Best Paper Award in Workshop on Symmetries in Robot Learning at RSS 2023.
[June 2023] Our paper "Robotic Manipulation Learning with Equivariant Descriptor Fields: Generative Modeling, Bi-equivariance, Steerability, and Locality" has been accepted to at RSS Workshop on Symmetries in Robot Learning 2023 (Oral). Check my presentation video here!
[Jan. 2023] Our paper "Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learning" has been accepted to ICLR 2023.