Minyoung Hwang

I'm a first year graduate student at MIT CSAIL, advised by Andreea Bobu. Previously, I worked at CMU, advised by Yonatan Bisk, and Allen Institute for AI (AI2). I received my M.S. and B.S. in ECE at Seoul National University (SNU).

Research Keywords: robotics, human-robot interaction, reinforcement learning, computer vision

Skills: I received numerous awards for talent in Mathematics. I was honored to be fully supported by the Presidential Science Scholarship in Mathematics during my undergraduate degree. For details, go to Honors and Awards. I TAed 4+ Math/ECE lectures and I love to share my knowledge and insights with others.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Twitter  /  Github

• 07/2024 I got accepted to the KEF(Kwanjeong Educational Foundation) and the SBS Foundation Scholarship for my graduate program.
• 06/2024 I'm joining MIT CSAIL as a graduate student for MS&PhD under the guidance of Prof. Andreea Bobu!
• 02/2024 Our paper Promptable Behaviors: Personalizing Multi-Objective Rewards from Human Preferences is accepted to CVPR 2024.
• 01/2024 I've started a research internship at the CMU Language Technologies Institute under the guidance of Prof. Yonatan Bisk!
• 10/2023 I received the NeurIPS 2023 Scholar Award for presenting in NeurIPS 2023.
• 09/2023 Our paper Sequential Preference Ranking for Efficient Reinforcement Learning from Human Feedback is accepted to NeurIPS 2023.
• 08/2023 I gave a talk at the CMU Language Technologies Institute about 'Leveraging Human Knowledge to Robots through Interaction: from Vision to Language and Beyond'.
• 07/2023 I got accepted to the Global Korea Scholarship for my prospective 2024 Fall PhD program.
• 07/2023 I've started a research internship at the Allen Institute for AI (AI2).
• 03/2023 I received the Google Student Travel Grant for presenting in CVPR 2023.
• 03/2023 Our paper Meta-Explore: Exploratory Hierarchical Vision-and-Language Navigation Using Scene Object Spectrum Grounding is accepted to CVPR 2023.
---- show more ----
10/2022 Our paper IMAGINE: Improving Image-Goal Navigation with Visual Language Grounding via Image Captioning is accepted to Conference on Information and Control Systems (CICS) 2022.
06/2022 We donated all royalties of <Wanna be Engineer> to the SNU College of Engineering. The donation will be used as a scholarship for high school students.
05/2022 Joined Sequor Robotics as a research manager for socially-aware navigation project.
01/2022 Our book <Wanna be Engineer> written by STEM members was selected as a best seller in Teen & Young Study Aids in Korea! I wrote about the joyfulness of developing robots.
10/2021 Our paper Geometric Understanding of Reward Function in Multi-Agent Visual Exploration is accepted to International Conference on Control, Automation and Systems (ICCAS) 2021.
01/2020 Won Grand Prize (1st Prize) at Silicon Valley Innovation and Startup Program! We developed fall detection and alert for elderly users.
10/2019 Invited to the Blended Science Festival to demonstrate motion controlled robot cleaner.
09/2019 Won 1st Prize at International Capstone Design Fair with our robot cleaner!
08/2019 Won 2nd Prize at SNU College of Engineering Creative Design Fair! We developed a Scrupulous Robot Cleaner with Mecanum Wheels and pointed-shaped inlet.
11/2018 Won Gold Medal at University Students Contest of Mathematics in Korea.

I'm interested in robotics, human-robot interaction, computer vision, and reinforcement learning. Much of my research is about applying mathematical concepts/theories to robot learning. Representative papers are highlighted.

Evaluating robot motions involves more than just the start and end states; it's about how the task is performed. We propose motion instruction fine-tuning (MotIF) and MotIF-1K dataset to improve VLMs' ability to understand nuanced robotic motions.

We propose a novel framework that facilitates efficient personalization of robotic agents to diverse human preferences in complex environments.

We propose a novel RLHF framework called SeqRank, that uses sequential preference ranking to enhance the feedback efficiency.

We propose and end-to-end learnable hierarchical navigation policy for vision-and-language navigation (VLN) that finds an unvisited and near-optimal local goal by recognizing semantically meaningful clues.

We propose a visual navigation method that the agent augments language observations and use the multi-modal observations for hierarchical exploration.

We present a new reward shaping method for multi-agent visual exploration with geometric understanding of the environment.

We propose a novel method to reconstruct 3D multi-view video with texture generation from a 2D human motion video.

I hosted and participated in some challenges.

We hosted RC car racing challenges in 2022 Spring & 2021 Fall ECE 430.457 'Introduction to Intelligent Systems'.

We present a startup item for fall detection for elderly using phone camera based on human pose estimation.

We present an omnidirectional robot cleaner with autonomous & hand-motion controlled modes. I leaded development of control algorithm considering robot dynamics.

We invented an envelope with built-in seive to sift powder automatically in the process of taking it out of the bag.

The first self help book written by SNU engineering students, including the stories of studying in middle and high school years, the reason for choosing engineering majors, and the career path after graduation.

I TAed the following courses in the department of Electrical and Computer Engineering and the department of Mathematics.

Managed RC car racing project and teach students how to use ROS and Gazebo. Head TA for 2022 Spring Lecture.

Tutoring for Seoul National University ECE major subject.

Supported and helped students do electric circuit experiments.

Taught Basic Calculus 1, 2 and Basic Calculus for Biological Science.

template adapted from this awesome website
Last updated: Jun 2024