Machine Learning Researcher

Imperial College London

J P Morgan Chase


Sanket Kamthe is a third-year PhD student at Imperial College London. He is focusing on reinforcement learning for robotics and control for his PhD. He is particularly interested in Safe Model-based Reinforcement Learning, where the agent learns to perform tasks while being aware of risks and uncertainties. He primarily works with Gaussian process models for uncertainty quantification.


  • Reinforcement Learning
  • Gaussian Processes
  • Probabilistic Modelling
  • Variational Inference


  • PhD in Computer Science, 2020

    Imperial College London

  • MRes in Advanced Computing, 2016

    Imperial College London

  • MSc in Information and Communication Engineering, 2014

    Technische Universität Darmstadt

  • B.Eng. in Electronics & Telecommunications, 2008

    University of Pune



Senior Associate

J P Morgan Chase

Sep 2020 – Present London, United Kingdom
Hardware software co design for multiplatform deployments of machine learning applications. Privacy preserving machine learning for business applications

Research Intern

J P Morgan Chase

Oct 2019 – Apr 2020 London, United Kingdom
Synthetic Data generation for privacy

Machine Learning Researcher


Apr 2018 – Sep 2019 Cambridge, United Kingdom
Data efficient reinforcement learning

Marie Curie Research Fellow

Dept. of Applied Mathematics, University of Twente

Apr 2014 – Sep 2015 Enscehde, Netherlands
Inference in high dimensional state space models.

Junior Research Fellow

Tata Institute of Fundamental Research

Mar 2010 – Aug 2011 Mumbai, India
Simulation, modelling and characterization of silicon based Single Photon Avalanche Detectors (SPAD) for CERN and TIFR, Ooty labs.

Project Engineer

Wipro Technologies

Nov 2008 – Feb 2010 Bangalore, India
Worked as a design engineer for Nortel Networks digital telephone systems. Managed a hardware product design cycle spread over 9 months and 3 continents from concept to field trials.

Recent Publications

Quickly discover relevant content by filtering publications.

Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control

Trial-and-error based reinforcement learning (RL) has seen rapid advancements in recent times, especially with the advent of deep …

Multi-modal filtering for non-linear estimation

Multi-modal densities appear frequently in time series and practical applications. However, they are not well represented by common …

Interaction primitives for human-robot cooperation tasks

To engage in cooperative activities with human partners, robots have to possess basic interactive abilities and skills. However, …