Anastasios Angelopoulos
Ph.D. student in Electrical Engineering and Computer Science
at the University of California, Berkeley
Student of Michael I. Jordan
and of Jitendra Malik
Research interests
Building out Distribution-Free Uncertainty Quantification, motivated by applications to medicine and computational imaging.
Recent News
August, 2022 Check out our paper, Conformal Risk Control. Shockingly, conformal prediction can be used to bound risk functions far beyond coverage. This is a short and useful read (the theory section is 2 pages).
July, 2022 Thanks to everyone who came to the 2022 ICML Workshop on Distribution-free Uncertainty Quantification on July 23, 2022. It was a blast!
July, 2022 I'm a bit late to shout out the following papers, so I'll do it all at once. NEW: Linear Revolution Invariance: Modeling and Deblurring Spatially-Varying Imaging Systems, Recommendation Systems with Distribution-Free Reliability Guarantees, Improving Trustworthiness of AI Disease Severity Rating in Medical Imaging with Ordinal Conformal Prediction Sets, Semantic uncertainty intervals for disentangled latent spaces. Also, Im2Im-Uq was accepted in ICML'22!
