Ruiqi Wang

/ Ray-chee Wong /

Hello!

I'm a Ph.D. candidate at Purdue University, where I work in the Smart Machine And Robotics Technology (SMART) Lab with Professor B.C. Min.

My research goal is to enable robots to integrate seamlessly into unstructured, human-centered environments.

To this end, I develop methods for robot learning from human feedback, enabling robots to acquire scalable, aligned, and personalized behaviors without hand-specified objectives. I pursue this goal along three directions:

Eliciting Informative Feedback

How can robots obtain informative feedback with minimal human effort?
My research uses foundation models as informative feedback priors to reduce human burden, while developing active, context-aware elicitation that decides when, what, and how to ask.

Learning Intent-Aligned Reward Representations

How can robots convert imperfect, multimodal feedback into reward representations that faithfully capture human intent?
My research develops reward learning methods that distill subtle, noisy, and multimodal feedback into structured representations of human intent, improving credit assignment and robustness to ambiguous, sparse, or conflicting supervision.

Sustaining Generalizable Alignment

How can robots remain aligned as tasks, users, and environments change?
My research develops transferable alignment mechanisms that generalize learned intent across tasks, users, and environments while preserving stable alignment, maintaining task performance, and avoiding drift from human intent.

Beyond learning from feedback, my research also studies how robots can respond to broader human signals in real time, including physiological state, cognitive workload, trust, and interaction dynamics, to sustain fluent, safe, and effective collaboration, especially in human-robot teaming.

Research Areas:

Learning from Human Feedback Preference-based Reinforcement Learning Human-Robot Interaction Multimodal Learning Foundation Models for Robotics Multi-Human Multi-Robot Teaming

News

2025
2024
2023
2022