Abstract
Lecture on cutting-edge HPC technologies for Deep Learning methods to address Earth Observation applications that deal with large and complex Remote Sensing datasets
General Information
MENELAOSᴺᵀ Midterm Summer School (MSS) 2021 is an event held by the MSCA ITN MENELAOSᴺᵀ in Santiago de Compostela (Spain). The Summer School is organized by CiTIUS, the Research Center on Intelligent Technologies of the University of Santiago de Compostela. The program consists of presentations and lectures about the newest achievements and developments in multimodal – multi sensor data fusion in the context of the MENELAOSᴺᵀ project.
Lecture
Deep Learning (DL) is emerging as the leading Artificial Intelligence (AI) technique owing to the current convergence of scalable computing capability. HPC systems are an effective solution to the challenges posed by big data analytics. In particular, applications which collect and store massive amounts of data can profit from them significantly. Modern Earth Observation (EO) programs, governments and space agencies are opening their archives(e.g., ESA’s Copernicus, NASA’s Landsat), making massive volumes of RS data available to everyone. Present and upcoming space missions use small satellite constellations to acquire data with higher spatial, spectral or temporal resolution. Other sources, such as ground and airborne sensors also provide a continuous stream of data. In this context, the main challenge is to identify efficient approaches to extract interpretable information and knowledge from this pool of data, possibly in near-real time and by integrating between different disciplines. This course will introduce students to cutting-edge HPC technologies for DL methods to address EO applications that deal with large and complex RS datasets.
