Understand the diverse applications of neural networks and deep learning in various fields such as text categorization, including spam filtering, fraud detection, optical character recognition; machine vision, including face detection and license plate recognition; advanced driver assistance systems; natural language processing, including spoken language understanding and market segmentation; and bioinformatics, including protein or lipidome classification based on function.Gain practical knowledge and skills in implementing neural network models and deep learning algorithms.Apply the acquired knowledge to solve classification problems in real-world scenarios, both for industrial and research purposes.Develop the ability to transfer the learned concepts and techniques to different domains and datasets.Analyze and evaluate the performance of neural network models and deep learning algorithms in classification tasks.Explore state-of-the-art use-cases and interesting applications of neural networks and deep learning, providing insight into current advancements in the field.By the end of this course, students will have a comprehensive understanding of neural networks and deep learning, enabling them to apply this knowledge to solve classification problems and explore cutting-edge applications in various domains.

the teaching methodology emphasizes a combination of short concept presentations and interactive programming sessions focused on relevant case studies. The use of eLearning tools enhances the learning experience, providing students with additional resources and opportunities for hands-on practice. These tools may include interactive simulations, online tutorials, and coding platforms, enabling students to actively engage with the course material and reinforce their understanding of neural networks and deep learning concepts. By incorporating both theoretical knowledge and practical application, the course promotes a comprehensive learning approach for mastering neural networks and deep learning techniques.

1. Parallel Computing & Setup:To provide a foundation for understanding the implementation of parallel systems, this lecture introduces parallel computing techniques and guides students through the setup process. It sets the stage for leveraging the power of multiple processors in neural network computations.2. Introduction to Machine Learning:With the parallel computing foundation in place, students delve into the fundamentals of machine learning. This lecture covers supervised and unsupervised learning, preparing them to explore how neural networks fit into the broader field of machine learning.3. Image Classification with Convolutional Neural Networks (CNNs):Students learn the principles of image classification using CNNs. They gain an understanding of the architecture of CNNs, their layers, training techniques, and their application in accurately classifying images into predefined categories.4. Object Detection:Continuing in the domain of computer vision, this lecture covers popular object detection algorithms, including Faster R-CNN and YOLO. Students learn how to use deep learning models to accurately and efficiently detect objects in images.5. Natural Language Processing (NLP) with RNNs:In this combined lecture, students explore the intersection of neural networks and natural language processing. They learn about recurrent neural networks (RNNs) and how they are applied in NLP tasks such as language modeling, sentiment analysis, and text generation.6. Transfer Learning and Fine-Tuning:Having gained a solid understanding of neural network fundamentals, students explore the powerful technique of transfer learning. This lecture covers how to leverage pre-trained models, freeze and fine-tune layers, and adapt existing models to new tasks and datasets.7. Generative Adversarial Networks (GANs):Introducing generative modeling, this lecture delves into GANs and their architecture. Students learn how GANs can generate realistic data, such as images, music, and text. They also understand the training process and challenges associated with GANs.8. Model Interpretability and Explainability:To round out the course, this lecture covers techniques for model interpretability and explainability. Students gain insights into analyzing and understanding the decision-making process of neural network models using methods like feature importance analysis and saliency maps.

The evaluation for the course "Lab: Neural Networks and Deep Learning" consists of two major components: Homework assignments and a final project presentation. The total grade is divided as follows:Homework Assignments: 40% of the total grade.Final Project Presentation: 60% of the total grade.

The course covers a range of topics related to Neural Networks and Deep Learning. The exam will focus on the following key areas:Fundamentals of Neural Networks: Understanding the basics of neural networks, activation functions, and architecture design.Deep Learning Techniques: Exploring various deep learning techniques such as Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs).Model Training and Optimization: Learning about optimization algorithms, loss functions, and strategies for training neural networks effectively.Real-world Applications: Applying neural networks to practical problems such as image classification, object detection, and image generation.Project Development: Designing, implementing, and presenting a deep learning project.

The final exam for this course is divided into two components: Homework assignments and a final project presentation.Homework Assignments(40%): The course includes four homework assignments, each contributing 10% to the overall grade. These assignments are designed to test your understanding of the course materials and your ability to apply the concepts learned.Final Project Presentation (60%): The final project accounts for a significant portion of your grade, comprising 60%. You will need to choose a Deep Learning task as your project and present it during the last two sessions of the class. The project presentation is structured as follows:Presentation (60pt): Prepare a  PowerPoint presentation with the following sections:IntroductionProblem DefinitionSystem Architecture (your proposed solution)Input -> Output samplesFuture ImprovementDemo of Model (20pt): Demonstrate the functionality of your model and its capabilities.Code (20pt): Share the code that you developed for your project.PowerPoint Presentation (Point Schema):Explain the problem your model is designed to solve. (10pt)Illustrate the layer order in the form of a diagram. (10pt)Describe the model's hyper-parameters, including optimization algorithm, batch size, and loss function.(10pt)Specify the dataset used for training, testing, and validation, along with the distribution of data among these sets.(10pt)Highlight the model's performance, indicating accuracy or error achieved.(10pt)Showcase some model outputs.(10pt)Provide suggestions for improving the model's accuracy.(10pt extra)Upload your project to Moodle before the last two sessions. Without uploading the project, the evaluation is not possible.Remember that your project presentation is a significant part of your grade and should demonstrate your understanding of the course concepts, your ability to implement deep learning models, and your creativity in solving real-world problems.