Presentations PDFs
Day 1 – 24/10 morning session
-
Context RAINSMORE and SWOT NE projects, and objectives of the workshop (Dr Marielle Gosset)
-
Time series forecasting using kernel regression methods. (Pr Guillerme Barreto)
Day 2 – 25/10 morning session
-
Time series prediction, from time alignment to basic algorithms - preprocessing (Dr Kelly Grassi)
-
Times series completion algorithms used for prediction - DTWB (Pr Emilie Caillault)
-
Times series prediction by LSTM. Theory and example from Ceara (Pr Nícolas Araujo’)
Day 3 – 26/10
-
Detection of patterns by Deep NN, transfert learning - (Pr Nicolas Araujo)
-
Training Rare Object Detection in Satellite Imagery by GAN and autoencoders (Pr Nicolas Araujo’)
Day 4 – 27/10
-
Clustering approach : Density, Shape, Hierarchical Kernel - approaches (Pr Emilie Caillault)
-
How to combine clustering and classification results, relevant object class (Pr Emilie Caillault’)
Day 5 – 28/10