Theo Bourdais

Theo Bourdais

PhD student in Computing and Mathematical Sciences

Posts by Collection

portfolio

publications

Development and Assessment of an Artificial Intelligence-Based Tool for Ptosis Measurement in Adult Myasthenia Gravis Patients Using Selfie Video Clips Recorded on Smartphones

Published in Digital Biomarkers, 2023

My first paper ! This paper describes the work we did at Doc.ai for ShareCare on Myasthenia gravis (MG) with UCB. We devellopped a Deep Learning model based on ResNet50 to predict a key symptom of MG via selfie. This would allow clinical trials to accept more patients as part of a study in order to scale it, as well as follow almost in real time the evolution of a patients symptoms using an app on their phone. Read more

Download here

Codiscovering graphical structure and functional relationships within data: A Gaussian Process framework for connecting the dots

Published in PNAS, 2023

This paper introduces Computational Hypergraph Discovery (CHD), a novel method for uncovering unknown functional relationships between variables within datasets, And represent them using a hypergraph. Read more

Recommended citation: Bourdais, T., Batlle, P., Yang, X., Baptista, R., Rouquette, N., & Owhadi, H. (2024). Codiscovering graphical structure and functional relationships within data: A Gaussian Process framework for connecting the dots. Proceedings of the National Academy of Sciences, 121(32), e2403449121. https://doi.org/10.1073/pnas.2403449121 https://www.pnas.org/doi/10.1073/pnas.2403449121

Model aggregation: minimizing empirical variance outperforms minimizing empirical error

Published in Arxiv, 2024

We present a general framework to combine existing models in a robust manner. Our method is flexible, places few assumptions on the models aggregated, and performs well in our experiments Read more

Recommended citation: Bourdais, T., & Owhadi, H. (2024). Model aggregation: minimizing empirical variance outperforms minimizing empirical error. arXiv [Cs.LG]. Retrieved from http://arxiv.org/abs/2409.17267 https://arxiv.org/abs/2409.17267

talks

Talk on Computational Hypergraph Discovery

Published:

My talk on Computational Hypergraph discovery! I have given versions of this talk at:

Abstract

Most scientific challenges can be framed into one of the following three levels of complexity of function approximation. Type 1: Approximate an unknown function given input/output data. Type 2: Consider a collection of variables and functions, some of which are unknown, indexed by the nodes and hyperedges of a hypergraph (a generalized graph where edges can connect more than two vertices). Given partial observations of the variables of the hypergraph (satisfying the functional dependencies imposed by its structure), approximate all the unobserved variables and unknown functions. Type 3: Expanding on Type 2, if the hypergraph structure itself is unknown, use partial observations of the variables of the hypergraph to discover its structure and approximate its unknown functions. While most Computational Science and Engineering and Scientific Machine Learning challenges can be framed as Type 1 and Type 2 problems, many scientific problems can only be categorized as Type 3. Despite their prevalence, these Type 3 challenges have been largely overlooked due to their inherent complexity. Although Gaussian Process (GP) methods are sometimes perceived as well-founded but old technology limited to Type 1 curve fitting, their scope has recently been expanded to Type 2 problems. We introduce an interpretable GP framework for Type 3 problems, targeting the data-driven discovery and completion of computational hypergraphs. Our approach is based on a kernel generalization of (1) Row Echelon Form reduction from linear systems to nonlinear ones and (2) variance-based analysis. Here, variables are linked via GPs, and those contributing to the highest data variance unveil the hypergraph’s structure. We illustrate the scope and efficiency of the proposed approach with applications to network discovery (gene pathways, chemical, and mechanical), and raw data analysis. Read more

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Heading 1

Heading 2

Heading 3

Read more

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.

Heading 1

Heading 2

Heading 3

Read more