The goal of [email protected] day is to gather faculty, researchers and students who work or are interested in applications of machine learning to science.
It is an opportunity for any of them to present their work in a poster session and hear about some of the exciting related work being done at EPFL. Students and young colleagues were particularly impacted by the COVID-19 pandemic and this day will re-create the atmosphere of conferences where one gets ample feedback on their work and has plenty of networking opportunities.
The event will provide an opportunity for a limited number of participants to present their work in talk form. You can apply for an oral contribution by submitting a title and an abstract when registering for the event.
The event will provide an opportunity for all participants to present their work in poster form to connect and discuss their research. You can apply for an oral contribution by submitting a poster title when registering for the event. This is strongly recommended but optional.
Format: Scientific poster – A0 (Portrait)
Print: Please note, that you will be in charge of printing the poster and bringing it on-site on June 23.
Audience Anyone is welcome to participate.
Price: Participation is free, but registration is required by June 18th, 2022. We are not accepting submissions for oral contributions anymore
9:00 – 9:05 Introduction to the AI4Science initiative
9:05 – 10:45 Talk session I
Alessandro Pau – Applications of Machine Learning in Nuclear Fusion Research
Puck van Gerwen – Metric Learning for Kernel Ridge Regression: Assessment of Molecular Similarity
Eniko Szekely – A direct and distributionally-robust approach to detection and attribution of climate change
Ilker Oguz – Programming Nonlinearities Inside Multimode Fibers for Efficient Machine Learning.
Indaco Biazzo – Epidemic inference problems: from contact tracing to phase diagrams
10:45 – 12:00 Poster session I + Coffee break
12:00 – 13:30 Lunch
13:30 – 14:50 Talk session II
Nataliya Lopanitsyna – Alchemical machine learning for high entropy alloys
Sauradeep Majumdar – Finding promising materials for energy applications through a diversity-driven search
Damiano Sgarbossa – Generative power of a protein language model trained on multiple sequence alignments
Lucien Krapp – Parameter-free geometric deep learning for accurate prediction of protein interfaces
14:50 – 16:30 Poster session II + Coffee break
16:30 – 17:50 Talk session III
Francesco Cagnetta – How deep convolutional networks learn hierarchical tasks
Jannes Nys – Embedding abelian and non-abelian symmetries in neural-network quantum states
Adam Gosztolai – Shaping up dynamical systems: using geometry to study the interplay of structure and dynamics
Guillaume Obozinski – Learning interpretable latent dynamics for a 2D airfoil system
17:50 – 19:00 Apéro
Note: This event is being photographed: Please be advised that CIS may use photographs taken during events for publication in print and online. Please alert the photographer or cis team member at the beginning of the event if you wish not to have your photograph taken.