My long-term goal is to build robots that continually use the knowledge gained during their lifespan to improve their performance and learn new tasks with human-like efficiency. As such, I work at the intersection of machine learning, robotics, and computer vision to develop intelligent robot systems with theoretically-founded lifelong learning capabilities. I am interested in a broad range of topics including Bayesian deep learning, conformal prediction, out-of-distribution generalization, and continual learning.
Mar 2, 2021
1 paper accepted to CVPR 2021 (Oral Presentation).
Jan 13, 2021
1 paper accepted to ICLR 2021.
Feb 1, 2020
1 paper accepted to ICRA 2020.
Nov 9, 2019
Presented my talk on how to evaluate predictive uncertainty estimates in deep object detectors at IROS 2019.