Quentin Berthet


Statistical Laboratory,
Department of Pure Mathematics
and Mathetical Statistics

University of Cambridge

I am a Lecturer in the Statslab, in the DPMMS at the University of Cambridge, and a fellow of St John's College. I am also a Turing Fellow at the Alan Turing Institute.

I completed my Ph.D. at Princeton University, under the supervision of Philippe Rigollet. I was also a CMI postdoc at Caltech, hosted by Venkat Chandrasekaran.

My interests are in Statistics, Optimization and Computer Science for Machine Learning, and the relationship between computational and statistical efficiency.

Publications & Preprints

Estimation of smooth densities in Wasserstein distance
with Jonathan Weed
Accepted at COLT 2019

Regularized Contextual Bandits
with Xavier Fontaine and Vianney Perchet

Statistical Windows in Testing for the Initial Distribution of a Reversible Markov Chain
with Varun Kanade

Unsupervised Alignment of Embeddings with Wasserstein Procrustes
with Edouard Grave and Armand Joulin

Detection of Planted Solutions for Flat Satisfiability Problems
with Jordan S. Ellenberg

Optimal Link Prediction with Matrix Logistic Regression
with Nicolai Baldin
Preprint, 2018

Fast Rates for Bandit Optimization with Upper-Confidence Frank-Wolfe
with Vianney Perchet
NeurIPS 2017, Spotlight presentation

Exact recovery in the Ising blockmodel
with Philippe Rigollet and Piyush Srivastava
Annals of Statistics, to appear, 2016

Average-case Hardness of RIP Certification
with Tengyao Wang and Yaniv Plan
NeurIPS 2016

Resource Allocation for Statistical Estimation
with Venkat Chandrasekaran
Proceedings of the IEEE, 104 (1), 111-125. 2016

Statistical and Computational Trade-offs in Estimation of Sparse Principal Components
with Tengyao Wang and Richard J. Samworth
Annals of Statistics, 44 (5), 1896-1930. 2016 Video

Optimal Testing for Planted Satisfiability Problems
Electronic Journal of Statistics, 9, 298-317. 2015

Computational Lower Bounds for Sparse PCA
with Philippe Rigollet
Extended version, Best Paper at COLT 2013, Video

Optimal Detection of Sparse Principal Components in High Dimension
with Philippe Rigollet
Annals of Statistics, 41(4), 1780-1815. 2013


2015: Lecturer, University of Cambridge
2015: Fellow, St John's College
2016: Faculty Fellow, Alan Turing Institute
2014-15: CMI Postdoctoral fellow, California Institute of Technology
2009-14: Ph.D. student, Princeton University
2006-09: Ecole Polytechnique, France
2004-06: Lycée Louis-Le-Grand, Paris, France
My resume: CV


Principles of Statistics (Part II, Michaelmas 2016, 2017, 2018)
Optimization (Part I, Easter 2016, 2018, 2019)
Topics in Statistical Theory (Part III, Lent 2016)