Achille Salaün

Achille Salaün

PhD in Computational Mathematics

Bonjour!

I am a graduate Télécom Paris engineer since 2017. Notably, I followed the EURECOM cursus where I specialised in Data Science from 2015-2017. In 2021, I received a PhD in computational mathematics (mathématiques aux interfaces in French) from the Institut Polytechnique de Paris. This was an industrial PhD (CIFRE) with Télécom SudParis (SAMOVAR) and Nokia Bell Labs supervised by , , , and . During that time, I focused on alarm prediction in telecommunication networks via space-time pattern matching and machine learning, aiming to provide experts with tools to understand how failures cascade across telecommunication networks.

Currently, I work as a postdoc at the Oxford Institute of Biomedical Engineering (part of the Department of Engineering of the University of Oxford), within the CHI Lab. With , , and , I develop interpretable AI tools to support clinical decisions in organ transplantation.

Feel free to reach out to me at .

Interests
  • Machine Learning
  • Interpretability
  • Pattern matching
  • Generative models
Education
  • PhD in Computational Mathematics, 2017-2021

    Institut Polytechnique de Paris

  • Master's degree in Data Science and Engineering, 2015-2017

    EURECOM

  • Diplôme d'ingénieur, 2014-2017

    Télécom Paris

Experiences

 
 
 
 
 
Explainable machine learning models for clinical decision support in kidney transplant offering
CHI Lab, IBME, Department of Engineering, University of Oxford
May 2022 – Present Oxford, United Kingdom

With at the CHI Lab, IBME, University of Oxford, and the clinicians and , I contribute to the construction of a computerised decision support system (CDSS) to help surgeons and patients understand if a given organ offer is a suitable one, or if it is preferable to wait for another one.

More precisely, given more than twenty years of data provided by the National Institute for Health and Care Research (NIHR) with ethical approval, we develop machine learning models predicting transplant outcomes (graft failure and patient death) in the case of an accepted offer, or telling what could happen in the case of a denied one. These models are supported by interpretability methods.

 
 
 
 
 
PhD: Alarm prediction in networks via space-time pattern matching and machine learning
Institut Polytechnique de Paris
Oct 2017 – Jul 2021 Paris, France

During this industrial PhD (CIFRE) between Télécom SudParis (SAMOVAR) and Nokia Bell Labs, I worked on predicting alarms in networks via space-time pattern matching and machine learning. For this, I benefited from the supervision of , , , and .

This thesis is two-fold. On the one hand, we proposed a structure, called DIG-DAG, able to store online causality chains observed within log. On the other hand, we compared analytically the expressivity of two popular generative models: Hidden Markov Models and Recurrent Neural Networks.

Apart from the theoretical work, my thesis shows a strong applied component. Indeed, Anne, Marc-Olivier and I implemented our DIG-DAG related algorithms into a Python 3 module, which has been at the core of a collaboration with Nokia’s business units.

Please follow this link for more details about my thesis.

 
 
 
 
 
Teaching
Télécom SudParis
Sep 2018 – Nov 2020 Evry, France
For two consecutive years, I supervised lab sessions for a course on Scientific Calculus (Master 1) and another one on Image Segmentation with Hidden Markov Models (Master 2) at Télécom SudParis. These courses were respectively coordinated by , and and .
 
 
 
 
 
Research Internship : Community Detection in Graphs
Télécom Paris
Mar 2017 – Sep 2017 Paris, France
This internship, supervised by , provided a great environment to play with Louvain algorithm and modularity.
 
 
 
 
 
Semester’s Project : generating cooking recipes thanks to Variational Autoencoders
EURECOM
Mar 2017 – Sep 2017 Sophia Antipolis, France
During this semester at EURECOM, I discovered Variational Autoencoders and Tensorflow under the supervision of who gave me the fun challenge to generate cooking recipes!

Coffee break ☕

The recent pandemic definitely taught us that social relationships are precious from both a personal and scientific point of view. Therefore, here are some topics I would be happy to discuss if we meet around a cup of coffee (it can be a glass of water as well)!

Naturally, we can chat about research. I would be glad to know more about your current works and answer any of your questions about mine (but you can already start by asking them there 👉 ).
In my spare time, I love drawing on my pen tablet: you can have a look on my creations there 👉 ! One of my dreams would be to draw my own comic book. Who knows, maybe one day?
I am fond of fencing, an activity that can be a sport, a game, and an art at the same time. When I'm not on the pistes, I can also enjoy a good run from time to time.