I am Linfeng Zhao (赵林风), a CS Ph.D. student at Khoury College of Computer Sciences of Northeastern University. I’m advised by Prof. Lawson L.S. Wong and also working with Prof. Robin Walters. My research interests lie in the intersection of machine learning, robotics and artificial intelligence.
I focus on decision-making side in robotics. To enable mobile-manipulation robots (e.g., Boston Dynamics Spot) to act in novel environments, I use planning (in symbolic and physical level) and learning (world models and RL control policies). A large body of my research involves how geometric/algebraic structure helps improve generalization and scalability, including: (1) symmetric representation learning of state and action, (2) symmetric/object-centric world modeling and planning, etc.
During PhD, I was a research intern at Boston Dynamics AI Institute with Jennifer L. Barry in 2023 Spring-Summer, and a research intern at Amazon Science with Kari Torkkola and Dhruv Madeka in 2021 Summer. During undergraduate, I interned at Microsoft Research Asia and earlier worked with Prof. Hao Su at UC San Diego.
We study E(2) Euclidean equivariance in navigation on geometric graphs and develop message passing network to solve it.
We study whether Euclidean symmetry can help in reinforcement learning and planning, which models the geometric transformations between reference frames of robots.
We formulate how differentiable planning algorithms can exploit inherent symmetry in path planning problems, named SymPlan, and propose practical algorithms.
We study how implicit differentiation helps scale up and improve convergence of differentiable planning algorithms.
We formulate compositional generalization in object-oriented world modeling, and propose a soft and efficient mechanism for practice.