Minjune Hwang

I am an incoming CS PhD student at the University of Southern California, where I will be working with Prof. Daniel Seita. Before moving to LA, I am currently spending time in Korea as a visiting scholar in Prof. Youngwoon Lee's lab at Yonsei.

I received a MS degree in Computer Science at Stanford University, where I worked on robot learning research with Prof. Fei-Fei Li and Prof. Jiajun Wu in the Stanford Vision Lab. Prior to that, I completed my undergraduate studies in Computer Science and Statistics from UC Berkeley.

I was also fortunate to tackle exciting research problems in Amazon Robotics, Microsoft Research, and Apple, where I developed novel algorithms and practical systems to solve various problems in robotics.

Email  /  Resume  /  GitHub  /  LinkedIn  /  Google Scholar

The goal of my research is to develop systems and learning algorithms for efficient robot manipulation, from various modalities of perception and human feedback. As of now, I am particularly interested in finding effective representations for learning challenging manipulation skills.

Specifically, my research focuses on methods for:

• Leveraging multimodal perception for challenging manipulation problems.
• Learning policies and modeling reward from human demonstrations [1] and feedback [2] for robot manipulation.
• Using symbolic [1] and hierarchical [2, 3] representations to tackle long-horizon robotic tasks.