Samuel Cohen

PhD in Machine Learning

University College London


I am a 2nd-year PhD student in the Statistical Machine Learning Group at UCL, advised by Professor Marc Deisenroth. Previously, I read the MSc in Statistical Science at Oxford University, advised by Professor Dino Sejdinovic where I worked on Optimal Transport. Prior to that, I studied the BSc in Mathematics at Imperial College London. My research interests include Optimal Transport, Reinforcement Learning, Gaussian Processes, Time Series and Generative Models.


  • Optimal Transport
  • Reinforcement/Imitation Learning
  • Gaussian Processes
  • Generative Models
  • Time Series


  • PhD in Machine Learning, 2019-2023

    University College London

  • MSc in Statistical Science, 2018-2019

    Oxford University

  • BSc in Mathematics, 2015-2018

    Imperial College London

Blog Posts

Estimating Barycenters of Measures in High Dimensions

Barycenters summarize populations of measures, but computing them does not scale to high dimensions with existing methods. We propose a …

Aligning Time Series on Incomparable Space

Data is often gathered sequentially in the form of a time series, which consists of sequences of data points observed at successive …

Healing Products of Gaussian Process Experts

Products of Gaussian process experts commonly suffer from poor performance when experts are weak. We propose aggregations and weighting …



Summer researcher

National University of Singapore (NUS)

Jul 2018 – Oct 2018 Singapore
  • Intern in Automatic Speach Recognition working on bilingual ASR using Deep Learning methods
  • Supervised by Pr. Haizhou Li and Dr. Emre Yilmaz 

Machine Learning research Intern

Check Point

Jun 2017 – Sep 2017 Israel
  • Developed an innovative machine learning framework for detecting new malware campaigns based on  mining suspicious graph morphism signatures in Android applications.
  • Developed a ML engine for Android Malware Detection that was implemented in production.

Data Science research intern


Jul 2016 – Aug 2016 Israel
  • Data science applied to queuing theory research project 
  • Built a queuing theory model on the cloud instance capacity needed by Ravello (recent acquisition).