I am a research scientist at ConsenSys and Imperial College London.
My current research at ConsenSys focuses on privacy preserving machine learning and decentralisation. I’m exploring the potential of bringing together techniques such as Federated Learning and Blockchains to serve improve our healthcare systems. Our group at ConsenSys is growing a strong expertise on Secure Computing (Trusted Execution Environments - TEEs, MPC, …) and Verifiable Computing (TEEs, Zero-Knowledge Proofs, …) with the goal to enhance machine learning deployments with these privacy preserving technologies.
At Imperial, I work in the BioMedIA group, under Prof. Daniel Rueckert, where I’m leveraging distributed computing for AutoML. We’re working on model selection, neural architecture search and hyperparameter tuning using a various range of methods. This is done in collaboration with colleagues at the Complex Systems Institute, Paris in the context of the OpenMOLE project.
PhD in High Performance Computing, 2013
Blaise Pascal University, France
MEng in Computer Science, 2010
BSc in Computer Science, 2007
Institut Universitaire et Technologique, Clermont-Ferrand, France
(A full list is available here: https://cv.archives-ouvertes.fr/jopasserat)
Wed, Sep 4, 2019,
Mon, Jul 10, 2017, ScalaX bytes meetup
Fri, Dec 9, 2016, Scala eXchange 2016
Donaco is an Imperial College London based startup that aims to facilitate online donations using machine learning.
OpenMOLE scientific platform, decentralised model exploration and parameter tuning on the iEx.ec network
Exploration, calibration and diagnosis of numerical model leveraging distributed computing environments.