Tutorial: Hands-on Introduction to Deep Learning

Coach: Raúl Benítez Iglesias

Universitat Politècnica de Catalunya

Abstract

The recent emergence of Deep Learning (DL) methods has triggered a change in paradigm in artificial intelligence and pattern recognition applications. In DL, an artificial neural network is used to extract deep features from the data therefore bypassing the need of a custom definition of meaningful features by an expert. In the course we will review the main elements and architectures commonly used in image classification and object segmentation. We will also cover important aspects such as transfer learning, data augmentation and interpretability methods. The course will be a hands-on introduction to DL in computer vision with step-by-step examples using the cloud Python programming platform google colaboratory http://colab.research.google.com/