POSTPONED Data Institute Seminar: Learning & Ocean Big Data

on the May 23, 2018

At 2:00 pm
Due to strikes, the seminar is POSTPONED. Ronan Fablet (IMT Atlantique, Lab-STICC/TOMS) will give a talk on Learning & Ocean Big Data: data-driven modelling, analysis, and reconstruction of complex dynamical systems.
Learning techniques and data-driven approaches are rapidly becoming relevant alternatives to classical modeldriven approaches for a large number of application domains, including for the study of phenomena governed by physical laws. They offer new means to take advantage of the potential of observation and / or simulation big data.
In this talk, we focus on the study of complex dynamic phenomena (e.g., trajectories, fluid dynamics) and in the definition of "data-based" stochastic representations of these dynamic phenomena. We illustrate the main underlying concepts and models, in particular analog and deep learning models, through different applications exploiting trajectory and imaging data in the field of ocean monitoring and surveillance (eg, interpolation of missing data, detection of abnormal behaviors, prediction).

Some references:

R. Lguensat, P. Tandeo, P. Aillot, R. Fablet. The Analog Data Assimilation. Monthly Weather Review, 2017.
R. Fablet, P. Viet, R. Lguensat. Data-driven Methods for Spatio-Temporal Interpolation of Sea Surface Temperature Images. IEEE Trans. on Computational Imaging, 2017.
R. Lguensat, P. Viet, M. Sun, G. Chen, F. Tenglin, B. Chapron, R. Fablet. Data-driven Interpolation of Sea Level Anomalies using Analog Data Assimilation. https://hal.archives-
S. Ouala, C. Herzet, R. Fablet. SST prediction and reconstruction using patch-level neural network representations. IEEE Int. Geoscience and Remote Sensing Symposium, IGARSS'2018, Valencia, July 2018.
D. Nguyen, R. Vadaine, G. Hajduch, R. Garello, and R. Fablet. A multi-task deep learning model for vessel monitoring using AIS streams. Submitted.

Published on May 23, 2018

Practical informations