I am a postdoc research associate at the Turing Institute, working as part of the Turing-Roche partnership on developing new statistical methods within conformal prediction that can be levelled for healthcare applications.
Previously, I obtained my PhD from the Laboratory of Statistics, Probability and Modélisation (LPSM) @ Sorbonne Université, under the supervision of Etienne Roquain, Tabea Rebafka and Nataliya Sokolovska; focusing on controlling prediction error in various ML tasks, using tools from multiple testing.
Research
Full list: Google scholar
Talks
- February 2025, Statistics Seminar @Université Paris-Dauphine (CEREMADE unit), Paris
- December 2024, International Seminar on Selective Inference (ISSI), online
- November 2024, Computational Seminar @immunai
- September 2024, Centre for AI Seminar @AstraZeneca
- June 2024, Workshop on heterogeneous and distributed data, Warwick University, Coventry
- November 2023, Workshop on Tests and Bandits, Potsdam
- November 2023, Workshop Stat. learning and AI for the analysis of the intestinal microbiota, Sorbonne Université, Paris
- October 2023, Workshop FAST-BIG, INRIA, Paris
- July 2023, JdS 23, Bruxelles
- June 2023, Reading group on conformal inference @ Institut de Mathématiques d’Orsay, Paris
- May 2023, Invited discussion @ International Seminar on Selective Inference (ISSI) on the talk “Selection by Prediction with (Weighted) Conformal p-values” by Ying Jin
- March 2023, LARGR Workshop, Lille
- February 2023, ASCAI Workshop, Munich
- September 2022, BASICS Workshop, Sorbonne Université, Paris
- August 2022, 12th MCP conference, Bremen
- July 2022, 3-day meeting of statisticians in Paris, IHP, Paris
- June 2022, Mathematical Methods of Modern Statistics 3, CIRM, Marseille
- June 2022, LPSM PhD students seminar, Sorbonne Université, Paris
- June 2022, JdS 22, Lyon
- May 2022, International Seminar on Selective Inference (ISSI), online
- March 2022, SansSouci Workshop, Toulouse
- November 2020, Journées EcoNet, online
Contact me at amarandon-carlhian at turing dot ac dot uk