Applied Deep Learning

This comprehensive textbook is designed for a 5th-year course in GI-IADS, offering an in-depth exploration into Neural Networks and Deep Learning. The content includes: A revisit of neural network fundamentals. Introduction to Deep Learning concepts. Exploration of forward and backward propagation techniques. Insight into optimization algorithms including Gradient Descent with Momentum, RMSprop, and Adam. In-depth look at parameters and hyperparameters. Techniques for hyperparameter adjustment and batch normalization. Understanding orthogonalization and metrics. Techniques for adjustment in learning, testing, and validation phases. Strategies to improve performance and prevent overfitting. Insights into transfer learning and multi-task learning. Introduction to TensorFlow and Keras, and basics of Pytorch. Industrial case studies for practical understanding and a final project for hands-on application. This textbook aims to provide a comprehensive understanding of the latest advancements in deep learning and their practical applications, particularly in industrial contexts.