IIHE invited seminar: Machine learning in scientific workflows

by Prof. Balázs Kegl (LAL)

Wednesday 2 May 2018, 14:00 → 15:00 Europe/Brussels

Jean Sacton Seminar room (1G003) (IIHE)

I will start by a walk through the sixty year history of machine learning with a focus on neural nets. I will shed light on the technological and algorithmic milestones that led to the current buzz on deep learning, and flash some of the major industrial applications that drive this current wave. In the second half of the talk I will describe the relevance of ML in sciences. I will detail some of the particular challenges of ML/science collaborations and the solutions we built to solve these challenges. I will illustrate this part by numerous scientific workflows that we built in the Paris-Saclay Center for Data Science (e.g., the Higgs boson discovery pipeline, El Nino forecasting, detecting Mars craters on satellite images, anomaly detection in particle physics detectors, classifying molecular spectra). Short bio: Balázs Kégl is a senior research scientist at CNRS and head of the Center for Data Science of the Université Paris-Saclay. He is co-creator of RAMP (www.ramp.studio).

Slides slides_kit_1804.pdf