Jekyll2026-01-30T19:00:26+00:00https://msesia.github.io/feed.xmlMatteo SesiaOfficial academic websiteMatteo Sesiasesia@marshall.usc.eduCOPA 20252025-07-10T00:00:00+00:002025-07-10T00:00:00+00:00https://msesia.github.io/posts/2025/07/10/COPA%202025Two of our papers received top honors this year at the 14th Symposium on Conformal and Probabilistic Prediction with Applications (COPA).

Best Paper: Conformal Survival Bands for Risk Screening under Right-Censoring, with Vladimir Svetnik.

Best Poster (one of two awards): Conformal Classification with New Labels, with my PhD student Tianmin Xie, Ziyi Liang, and Stefano Favaro.

]]>Matteo Sesiasesia@marshall.usc.eduGoogle Research Scholar Award2025-05-30T00:00:00+00:002025-05-30T00:00:00+00:00https://msesia.github.io/posts/2025/05/30/Google%20AwardMy research proposal on conformal inference for uncertainty estimation in real-world machine learning has been generously selected for funding by Google Research. Thank you!

]]>Matteo Sesiasesia@marshall.usc.eduICML 20252025-05-01T00:00:00+00:002025-05-01T00:00:00+00:00https://msesia.github.io/posts/2025/05/01/ICML%202025My 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

]]>Matteo Sesiasesia@marshall.usc.edu