Dartmouth Events

Abstract:

Is there a unified approach to achieving robot autonomy? In this talk, Dr. Wang will address this question by introducing Imperative Learning (IL), a self-supervised neuro-symbolic reasoning framework. IL is formulated as a specialized bilevel optimization (BLO) framework, integrating three core components: a neural module, a reasoning engine, and a memory system. This architecture enables reciprocal learning across these components, leveraging the expressiveness of neural models while maintaining the interpretability of symbolic reasoning. Dr. Wang will demonstrate how IL can be applied to a range of robot autonomy tasks, including localization and mapping, control and planning, rule induction, off-road driving, and vision-language navigation. Through extensive experiments, IL has been shown to significantly enhance autonomous capabilities, offering a promising direction for future research in robotics and AI.

Bio:

Dr. Chen Wang is an Assistant Professor and the Director of the Spatial AI & Robotics (SAIR) Lab (https://sairlab.org) at the Department of Computer Science and Engineering (CSE), University at Buffalo (UB). He received his B.Eng. from Beijing Institute of Technology (BIT) in 2014 and his Ph.D. from Nanyang Technological University (NTU), Singapore, in 2019. From 2019 to 2022, he was a Postdoctoral Fellow at the Robotics Institute, Carnegie Mellon University (CMU). Dr. Wang's research focuses on advancing human-level spatial awareness and reasoning in robotic systems. He is the creator of PyPose (https://pypose.org), a widely used library for spatial AI applications, which has garnered over 5,000 monthly downloads. He serves as Associate Co-Chair for the IEEE Technical Committee on Computer & Robot Vision and as an Associate Editor for leading robotics journals, including IJRR and RA-L. Additionally, he has served as an Area Chair for top-tier conferences such as ICRA, CVPR, and NeurIPS. Dr. Wang has received several prestigious awards, including the Best Paper Award in Robotic Planning at the International Conference on Advanced Robotics (2017), the Cisco Research Award (2023), the Sony Faculty Innovation Award (2024), and the UB CSE Excellence in Research Award (2024).