SIAM AI in Healthcare Event 13th May

Do you want to work in data-driven healthcare research? Are you curious how AI influences the healthcare industry?

Join us on Tuesday, May 13th, starting from 5 pm in room SG 1 for a workshop on career paths and research insights, followed by a networking apéro. Please register via the QR code below or under go.epfl.ch/siam-ai.

We are happy to welcome Kostas Sechidis from Novartis, Sebastien Benzekry from INRIA’s INSERM COMPO group, and Vincent Stimper from Isomorphic Labs. Please find excerpts of their bios below.


Kostas Sechidis is Associate Director of Data Science and research scientist at Novartis where he specializes in exploratory analysis for biomarker discovery, digital biomarker development, and assessing treatment effect heterogeneity in clinical trials. He did his PhD at the University of Manchester and subsequently worked as a postdoctoral researcher with AstraZeneca and Roche.

Sebastien Benzekry is a research scientist at Inria where he leads the INSERM project team Computational Pharmacology and Clinical Oncology (COMPO) modelling data arising from oncology to enhance therapeutic management, inform and design clinical trials, and test biological hypotheses. He did his PhD at Aix-Marseille University followed by a PostDoc at the Center for Cancer and Systems Biology, Tufts Univ, Boston, MA.

Vincent Stimper is a Machine Learning Research Scientist at the Google DeepMind’s sister company Isomorphic Labs where they use machine learning to develop novel drugs for challenging diseases. Before joining Isomorphic Labs in 2024 he did his PhD at the University of Cambridge and the Max Planck Institute for Intelligent Systems and completed internships at Microsoft, Amazon and the Ontario Institute for Cancer Research.

Jane Street virtual Estimathon, October 24th

Tomorrow, October 24th at 18:00 CET, will take place the virtual “Estimathon” hosted by Jane Street: 13 unique estimation problems in 30 minutes, combining maths and trivia, to be solved as a team.

It’s good fun, a good challenge, and a great opportunity to learn more about Jane Street! Sign up here or follow the poster below to obtain the Zoom link.

SIAM Professional Journey Showcase 2024: Dirk Hartmann

The Professional Journey Showcase took place this Tuesday, October 8th. Thanks to everyone who attended, and special thanks to Dirk Hartmann from Siemens for an engaging presentation and answering our questions!

Thanks as well to the EPFL mathematics department and Agepoly for their financial support.

G-Research Quant Finance Workshop, Wednesday 16th October

Are you an EPFL PhD student and curious about the world of Quantitative Finance?
You are invited to participate in G-Research’s Interactive Quant Finance Workshop on Wednesday, Oct. 16, from 18:00 to 22:00 at EPFL room GC B3 31.

🔍 What to Expect: You will take part G-Research’s brand new cointegration challenge, where you’ll test your strategies on unseen, real world datasets. Play for fun, or bring your competitive spirit for the chance to win prizes! Following the challenge you’ll have the chance to network with G-Research’s researchers over some food and drink!

👥 The workshop is a challenge that can be enjoyed solo, or in teams of up to three, so feel free to share the registration link with your friends and colleagues.

👉 Sign up now: qrco.de/EPFL2024 (or the QR code in the poster below)

SIAM Professional Journey Showcase 2024: Dirk Hartmann

Interested in a career in Applied Mathematics, Computational Science or Data Science, but still have some doubts?

Join us on Tuesday, October 8th, for the Professional Journey Showcase with Dirk Hartmann! The event will take place at EPFL in room MA B1 11, starting from 17h.

The workshop will consist in a short presentation, a career Q&A, and will be followed by a networking apéro. Registration is recommended (this link, or the QR code in the poster below).

Dirk Hartmann is an industrial mathematician, intrapreneur, and thought leader in the field of Simulation and Digital Twin. In his career, he took several leading roles in research, innovation, and development across Siemens including the lead of a multi-million Siemens R&D program and the technical leadership for the Simulation & Digital Twin field at Siemens Technology with more than 120 scientists across the globe. Many of his innovations have led to novel products and are regularly showcased at Siemens innovation events. For these achievements, he has received numerous Siemens awards, including the prestigious Werner-von-Siemens Top Innovator Award in 2019 and the Siemens Inventor of the Year Award in 2021.

Scimpact 2024-2025

This year too, we are happy to relay the announcement of the program Scimpact organized by Reatch, which may be of interest to some students at EPFL:

You want to learn how to write a good blog-article or moderate a discussion? You want to be part of a young science community that wants to make a difference? Welcome to Scimpact! Scimpact is a training program for young people who want to bring science into societal debates. The program consists of hands-on workshops and 1:1 coaching, lasts 4 or 8 months and offers you the chance to organize a public event! Apply by September 30 at www.reatch.ch/en/scimpact

Online Seminar: Learning Solution Operators for PDEs with Uncertainty

Join us for an online seminar on Monday, June 10th at 4PM, given by Emilia Magnani, a Ph.D. candidate at the University of Tübingen. She will present her work on “Learning Solution Operators for PDEs with Uncertainty”.

Abstract: We provide a Bayesian formulation of the problem of learning solution operators of PDEs in the formalism of Gaussian processes. We consider neural operators, recent deep architectures that have shown promising results in tackling the task of learning PDE solution operators. The current state of the art for these models lacks explicit uncertainty quantification. Our approach offers a practical and theoretically sound way to apply the linearized Laplace approximation to neural operators to provide uncertainty estimates. Moreover, we introduce a new framework for Bayesian uncertainty quantification in neural operators using function-valued Gaussian processes.

Bio: Emilia Magnani is a Ph.D. candidate at the University of Tübingen under the supervision of Philipp Hennig. She is also part of the ELLIS program and spent part of her Ph.D. in Genoa working with Lorenzo Rosasco. Before that, Emilia obtained her Master’s degree in Mathematics from ETH Zurich. Her research interests span various areas of machine learning such as probabilistic numerics, Gaussian processes, and operator learning.

Zoom Link: https://epfl.zoom.us/j/63925499984?pwd=GOEI1rAQrMOXaIgFLG5B3IYle4Funr.1

Online seminar on optimal sampling

This Wednesday (April 17th), we are pleased to host a talk by Philippe Trunschke (PostDoc at Centrale Nantes & Nantes Université), which should be interesting to many of our members!

The seminar will take place online at the zoom link below, and will also be projected live in the room CM 1 517. https://epfl.zoom.us/j/61353461236?pwd=MnI2VkRMWlE2WUJxalRmNVJwc2JGQT09

Title: Optimal sampling for stochastic gradient descent

Abstract: Approximating high-dimensional functions often requires optimising a loss functional that can be represented as an expected value. When computing this expectation is unfeasible, a common approach is to replace the exact loss with a Monte Carlo estimate before employing a standard gradient descent scheme. This results in the well-known stochastic gradient descent method. However, using an estimated loss instead of the true loss can result in a “generalisation error”. Rigorous bounds for this error usually require strong compactness and Lipschitz continuity assumptions while providing a very slow decay with increasing sample size. This slow decay is unfavourable in settings where high accuracy is required or sample creation is costly. To address this issue, we propose a new approach that involves empirically (quasi-)projecting the gradient of the true loss onto local linearisations of the model class through an optimal weighted least squares method. The resulting optimisation scheme converges almost surely to a stationary point of the true loss, and we investigate its convergence rate.

Philipp TRUNSCHKE studied Mathematics at the Humboldt University in Berlin, specialising in statistical learning theory. He completed his doctoral studies focusing on tensor product approximation at the Technical University of Berlin 2018. Currently, he is working with Anthony NOUY in Nantes on compositional function networks and optimal sampling.