I am a Florence Nightingale Bicentennial Fellow and Tutor in Computational Statistics and Machine Learning at the University of Oxford. Prior to that, I was a Data Science Initiative Postdoctoral Fellow at Harvard University, working in the lab of Pierre Jacob. Previously, I did my PhD in Statistics at Columbia University mentored by Liam Paninski, where I developed algorithms for extracting neural signals from recordings.
I develop robust, efficient and theoretically sound statistical methodology for tackling problems in life sciences that involve data. My experience is mostly in neuroscience and epidemiology, but I am open to other problems as well. On a more theoretical side, I am interested in enhancing our understanding of the statistical and computational aspects of entropic optimal transport, so new statistical procedures and algorithms can be derived. Finally, from a machine learning perspective, I am interested in the representation of complex combinatorial objects, such as permutations.
- May 9th, 2021. I will present my work on inference of stratified infection fatality rates in the End-to-end Bayesian learning international conference. This conference will take place 25-29th October 2021, in Marseille, France.
- April 28th,2021 our work on the impact of COVID-19 in Santiago, Chile has been published in Science. This is joint work with Pamela Martinez (co-first author), Ayesha Mahmud, Pablo Marquet, Caroline Buckee and Mauricio Santillana.
- Jan 07th,2021 our work on neural identification in colorful C.elegans worms has been published in Cell. This is joint work with Eviatar Yemini, Albert Lin, Amin Nejatbakhsh, Erdem Varol, Ruoxi Sun, Aravinthan DT Samuel, Liam Paninski, Vivek Venkatachalam and Oliver Hobert
- July 7th,2020 I gave a talk at Optimal Transport: Regularization and Applications Workshop.
- July 1st,2020 Started a new position as I am a Florence Nightingale Bicentennial Fellow and Tutor in Computational Statistics and Machine Learning at University of Oxford.
- June 30th, our new work on Sinkhorn EM algorithm (joint with Amin Nejatbakhsh, Erdem Varol, Jonathan Niles-Weed) is available as a preprint.