Artificial Intelligence I

Module 1 This module introduces fundamental concepts and terminology in artificial intelligence and machine learning, with a focus on supervised learning (regression analysis, classification), unsupervised learning, and reinforcement learning. The module systematically covers the entire AI/ML workflow: data collection, preprocessing, visualisation, modeling, and evaluation. Classical algorithms such as decision trees, random forests, k-nearest neighbors (KNN), linear models, clustering algorithms, and ensemble methods are central. Students learn techniques for model evaluation, including cross-validation and regularisation. Handson lab sessions using libraries such as Scikit-learn provide opportunities to implement these models in order to solve real-world problems involving various data types. The module also equips students with the skills to independently analyse datasets, select appropriate algorithms, and implement solutions using standard machine learning tools. Module 2 This module focuses on the theoretical foundations and practical applications of deep learning. Key principles include neural network architectures, the universal approximation theorem, and optimisation via gradient descent. Students learn to work with popular models such as convolutional neural networks (CNNs), ResNets, and generative models (e.g., autoencoders and language models). The module integrates hands-on exercises using libraries like PyTorch to adapt and apply pretrained models. Emphasis is placed on optimisation techniques such as transfer learning and data augmentation, as well as evaluating model performance using appropriate metrics. The module also introduces fundamental concepts in generative AI, including visual models and language models.