Artificial Intelligence (AI) is everywhere: in our cell phones as well as in our cars, both autonomous and non-autonomous. In fact, thanks to neural networks, AI is now able to understand much of the world around us.<\/p>\n
Neural networks, a computational mechanism inspired by the human brain, can learn complex tasks from examples. Today, neural networks are the most powerful artificial intelligence models for understanding images, text, and sound.<\/p>\n
In this course, we will focus on images, introducing computer vision and AI techniques for image classification, the simplest but perhaps most significant visual task.<\/p>\n
The course is divided into three modules and introduces the use of neural networks for the classification of tabular data and images. The training program combines traditional lectures (slides, whiteboard) with hands-on laboratory activities, giving participants direct programming experience.<\/p>\n
The first module introduces the fundamental concepts of Python programming, which will be used to visualize and manipulate digital images and perform simple data preparation operations for training neural networks.<\/p>\n
The second module is dedicated to classification and introduces neural networks, which will be used in a laboratory session for image classification.<\/p>\n
The third module introduces Convolutional Neural Networks (CNNs), the most widely used architecture in deep learning. CNNs will be used to successfully tackle more complex image classification problems.<\/p>\n
During the afternoon practical sessions, students will learn step-by-step how to process images with a computer and how to program a neural network to perform image classification.<\/p>\n
Classification and Neural Networks:<\/strong> The classification problem, from perceptrons to feed-forward neural networks, network training and performance evaluation.<\/p>\n Image Processing Fundamentals:<\/strong> Images and their representation, basic image manipulation, convolution and morphological operations for feature extraction.<\/p>\n Classification and Neural Networks:<\/strong> The classification problem, from perceptrons to feed-forward neural networks, network training and performance evaluation.<\/p>\n Deep Learning and CNNs:<\/strong> the Deep Learning revolution, Convolutional Neural Networks (CNNs), CNN training and performance evaluation.<\/p>\n Deep Learning and CNNs:<\/strong> the Deep Learning revolution, Convolutional Neural Networks (CNNs), CNN training and performance evaluation.<\/p>\n Video presentation<\/strong><\/p>\n <\/p>\n
\n