I am a technology enthusiast with an entrepreneurial mindset. I have co-founded Eliot, a data-driven start-up that helps people evaluate short-term rental opportunities, and was acquired by Airbnb. Prior to that, I have conducted research in machine learning at Stanford and MIT, and received a master’s degree from Ecole Polytechnique and Stanford (ICME).
If you find something interesting to you here, feel free to reach out !
One of my main focus at the Shah Lab was to design data-driven models to learn effective, low-dimensional representations of clinical data. I present in this post the common challenges that working with clinical data raises and two different approaches to representation learning and transfer learning.
The goal of this project was to implement AI-based strategies for a tailor-made game of multiplayer snake, inspired by the popular online game slither.io. The main motivation was to assess the relative performance of adversarial strategies versus reinforcement learning and compare snake behaviors inherent to each. This game setting comprises a number of challenges, such as simultaneous player actions, multiple opponents and large state spaces. Most of all, there is no single conspicuous objective, making it difficult to predict the opponents’ best moves in search trees and makes RL policies very dependent on the strategies used for training.
Eliot - www.eliotandme.com
MS in Data Science at ICME and research assistant at BMIR
Research Internship at Alfa Group
Co-founded a student association at Ecole Polytechnique to develop entrepreneurial activities
Diplome d’Ingenieur (MS) in Computer Science and Applied Mathematics
Feel free to contact me by email.