Alexis Laignelet

I am a PhD student in the Department of Computing at Imperial College London, where I work on smoothing techniques and implicit gradient methods. My PhD supervisor is Panos Parpas. I also work with Nikolas Kantas and Grigorios A. Pavliotis from the Department of Mathematics. This follows a MSc in Machine Learning from Imperial College London.

From 2014 to 2018, I worked at Eiffage, as a Structural Engineer. I used to design bridges and large factories. I also worked on time series datasets of the monitoring of Millau's Viaduc.

I have a MSc in Civil Engineering from Ponts ParisTech. I also have a MSc in Applied Mathematics from Mines ParisTech.

Email  /  GitHub  /  Google Scholar  /  LinkedIn  /  CV

Research

Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations

In this project we discuss the dynamical systems point of view of deep learning to solve PDEs and compare several architectures in terms of stability, generalization, and robustness. In order to speed up the computations, we propose to use a multilevel discretization technique.

Combination of multiple Deep Learning architectures for Offensive Language Detection in Tweets

The project aims at developing an architecture that detects offensive Tweets. This is part of a challenge: OffensEval 2019, and more specifically SemEval 2019 - Task 6. The competition was based on the Offensive Language Identification Dataset.

Talks

Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations

Based on the research project ‘Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations’

Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations

Based on the research project ‘Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations’

Multi-armed bandits

A review of the most common forms of multi-armed bandits with some practical examples.

Design and source code from Jon Barron's website