Assistant Professor Harold Soh Soon Hong
ACCEPTING PHD STUDENTS
Faculty & Department
Joint Appointments
Assistant Professor, Ssi Hq, Integrative Sciences and Engineering
Jt Appt - Asst Prof, SSI HQ, Smart Systems Institute
Education
Doctor of Philosophy, Imperial College London, United Kingdom
Bachelor of Arts & Science, University of California,Davis, United States
Bio
Harold Soh is an Assistant Professor of Computer Science at the National University of Singapore, where he leads the Collaborative Learning and Adaptive Robots (CLeAR) lab. He completed his Ph.D. at Imperial College London, focusing on online learning for assistive robots. Harold’s research interests are in machine learning, particularly generative models, and decision-making for trustworthy collaborative robots. His contributions have been recognized with a R:SS Early Career Spotlight in 2023, best paper awards at IROS’21 and T-AFFC’21, and several nominations (R:SS’18, HRI’18, RecSys’18, IROS’12). Harold has undertaken significant roles in the HRI community, most recently as co-Program Chair of ACM/IEEE HRI’24. He is an Associate Editor for the ACM Transactions on Human Robot Interaction, Robotics Automation and Letters (RA-L), and the International Journal on Robotics Research (IJRR). He is a Principal Investigator at the Smart Systems Institute and a co-founder of TacnIQ, a startup developing touch-enabled intelligence.
Contact Information
At CLeAR, we seek to improve people’s lives through intelligent robotics. We advance the science and engineering of collaborative robots that fluently interact with us to perform tasks. Our central focus has been on developing physical and social skills for robots. For the former, we’re working on new tactile perception and control methods for robots. In the latter, we’re developing better human trust models and social-projection-based communication.
A final third research thread is dedicated towards machine-learning research, particularly in robot learning and generative ML/AI. For example, we have devised novel methods for regularizing deep networks with symbolic knowledge, which we later showed improved robot imitation learning for a cooking task. Our work on sample refinement using gradient flows provides us a way to meld physical and social skills for tasks such as semantic grasping. These methods not only contribute to the wider machine-learning literature, but form a unique suite of methods CLeAR uses to advance the state-of-the-art on trustworthy collaborative robots
For more information, please see our CLeAR lab website.