Matteo Sesia

Statistician and Data Scientist. Associate Professor at the University of Southern California.

My collaborators and I had two papers accepted to the International Conference on Machine Learning (ICML) this year, one of which was selected as a spotlight (top 2.6% of submissions).

The poster paper, with Meshi Bashari and Yaniv Romano, introduces a robust conformal outlier detection method that can work with contaminated reference data, using an active data-cleaning strategy. Read more

The spotlight paper, with Vladimir Svetnik, presents a doubly robust conformal method for survival analysis with right-censored data, offering reliable inferences even when survival models are misspecified. Read more