GIACOMO BORACCHI - TEACHING
 

Advances Deep Learning  


AA 2018/2019, PhD Course, Politecnico di Milano


Mission:
The course presents a general and unified view over most successful architectural patterns in deep neural networks, and the machine learning problems they are naturally suited for. Each lecture will be organized around one of the most relevant deep learning architectures or learning problems (described below).
More information on the Course program page


Teaching Materials:

Fisrt Class: Convolutional Neural Networks for advanced visual recognition tasks Convolutional neural networks, fully convolutional networks, semantic segmentation. Second Class: Unsupervised Models Class-activation mapping, Denoising through Deep-Learning models, Remarks on the receptive field.