Spain

In this presentation I give an overview of the progress in H2020 project ExtremeEarth (http://earthanalytics.eu/) which investigates scalable big data and deep learning techniques for Copernicus data. The project is driven by two use cases in the areas of Food Security and Polar Regions.The former use case aims to develop irrigation recommendations from farmers while the latter will produce ice maps that can aid maritime users in the Arctic. The contributions of the project in the area of deep learning are twofold: (i) the production of two very large training datasets (millions of samples) for crop type mapping and ice classification, and (ii) the study of LSTM and CNN architectures for the same problems. In the area of big data, the project is developing four big linked geospatial data systems and a benchmarking environment for evaluating their performance and scalability. The development of all algorithms, systems and use cases takes place in the big data and AI platform Hopsworks (https://www.logicalclocks.com/) and run on CREODIAS.

ECAI 2020

Scalable big data and deep learning techniques for Copernicus data

ECAI2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI, including:




**An extensive programme of workshops, tutorials, and invited speakers**

**The Prestigious Applications of Intelligent Systems conference (PAIS)**

**The Frontiers in AI track sessions**

**The Lab To Market event**

**The Women in AI time**

**The EU Challenges forum**

… and many more!

ECAI 2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI.

technical paper

Equal Pay for Equal Competences : A Statistical  Approach to Address Equal Pay Gap

On Minimizing Diagonal Block-wise Differences for  Neural Network Compression

Deep Density-aware Count Regressor

DeepGS: Deep Representation Learning of Graphs and Sequences for Drug-Target Binding Affinity Prediction

Robot self other distinction active inference meets neural networks learning in a mirror

Minimality of Combined Qualitative Constraint Networks

The combination of logic and probability is very useful for modeling domains with complex and uncertain relationships among entities. Machine learning approaches based on such combinations have recently achieved important results, originating the fields of Statistical Relational Learning, Probabilistic Inductive Logic Programming and, more generally, Statistical Relational Artificial Intelligence.
The tutorial will concentrate on Probabilistic Logic Programming, a form of Probabilistic Programming that is receiving an increasing attention for its ability to combine powerful knowledge representation with Turing completeness.

This tutorial will introduce probabilistic logic programming and overview the main systems for learning models in these formalisms both in terms of parameters and of structure. The tutorial includes a significant hands-on experience with the systems ProbLog2 and cplint using the web applications https://dtai.cs.kuleuven.be/problog/ and http://cplint.eu that the attendants can access with their notebooks via wifi.


Probabilistic Inductive Logic Programming

Towards an Explainable Argumentation-based Agent

Differential evolution (DE) is a population-based stochastic parallel optimization algorithm that has been successfully applied to solve many optimization problems in Engineering. It is well known that DE performance depends strongly on its control parameters such as mutation factor, crossover probability, and population size. Although the former two control parameters have received a lot of attention, the later has not received the attention it deserves. Like many evolutionary algorithms (EAs), the population size in DE is important as it influences whether good solutions can be found and the time it takes to find them. Generally, the population size of DE algorithm is specified by the user and remains fixed during the run. Therefore, inappropriate choice of the population size may severely hinder the performance of DE algorithm. Previously, DE population size was treated as problem independent. However, recent research suggested that population size is problem dependent; in particular when dealing with high dimension problems. This paper investigates the impact of the population size on the performance of DE when applied to the optimal tuning of power system stabilizers (PSSs). Several population sizes have been investigated and their impact on the performance of the PSSs are presented. The effectiveness of these controllers is assessed using modal analysis and validated by time domain simulations.

Impact of Population Size on the Performance of DE Based Power System Stabilizers

"We propose a Generative Adversarial Networks (GANs) based architecture for cloud removal from satellite imagery. Data used for training comprises of visible light RGB and near-infrared (NIR) band images. The novelty lies in the structure of discriminator in GANs architecture. We propose the discriminator to compare generated and target cloud-free RGB images concatenated with their edge-filtered versions. The motivation behind this approach is that such a discriminator drives the
learning of the generator towards better identification of objects
covered by clouds. In the experimental results, we compare our
method with the benchmark solution, and demonstrate that our approach is superior in performing cloud removal for various cloud intensity levels."

Scalable big data and deep learning techniques for Copernicus data

Downloads

Next from ECAI 2020

Equal Pay for Equal Competences : A Statistical Approach to Address Equal Pay Gap

Similar lecture

Transcoding-based self-supervised learning for semantic segmentation of PolSAR imagery

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Scalable big data and deep learning techniques for Copernicus data

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ECAI 2020

Equal Pay for Equal Competences : A Statistical Approach to Address Equal Pay Gap

Similar lecture

Transcoding-based self-supervised learning for semantic segmentation of PolSAR imagery

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads