Yating Lin

I began my research in robot learning at the University of Michigan. Prior to that, I conducted research on optical microcavities at Zhejiang University. I got a bachelor's degree in Optoelectronic Information Science and Engineering from Chongqing University.

Email / Scholar / Github

Research

I'm interested in out-of-distribution generalization and detection for robot learning. And I'm also curious about how to leverage the representation learning and causality in generative models to achieve more robust and better generalized models for robotics.

	AnoF-Diff: One-Step Diffusion-Based Anomaly Detection for Forceful Tool Use Yating Lin, Zixuan Huang, Fan Yang, Dmitry Berenson paper We propose a parallel one-step diffusion approach for real-time anomaly detection in forceful tool use.
	Implicit Contact Diffuser: Sequential Contact Reasoning with Latent Point Cloud Diffusion Zixuan Huang, Yinong He, Yating Lin, Dmitry Berenson paper We enable the robot to reason about the changing contacts between objects and environment learning a diffusion model over neural contact representations.
	Subgoal Diffuser: Coarse-to-fine Subgoal Generation to Guide Model Predictive Control for Robot Manipulation Zixuan Huang, Yating Lin, Fan Yang, Dmitry Berenson paper We use diffusion model to generate subgoals at appropriate temporal resolution dynamically to guide MPC.
	Improving Out-of-Distribution Generalization of Learned Dynamics by Learning Pseudometrics and Constraint Manifolds Yating Lin, Glen Chou, Dmitry Berenson paper We leverage sparsity and nonholonomic constraints to improve out-of-distribution generalization in dynamics learning.
	Fiber-integrated open Fabry-Perot microcavity for high-brightness single-photon sources Coming soon We realize fiber-based Fabry–Perot microcavities with coupling efficiencies up to 96.3%, providing a platform for Purcell enhancement toward single-photon sources.

Miscellanea

I like to play tennis! Therefore, I'm also interested in sports and agile robots!

Physics and machine learning are both attractive to me, maybe someday learning for optics too!