Introduction to Machine Learning

Lectures

• Artificial Intelligence and Machine Learning
  Date: May 17, 2025 Description: This lecture presents the Artificial Intelligence and Machine Learning concepts. Their definition, history, implications, and applications.
• The Problem First
  Date: May 24, 2025 Description: This lecture will emphasis on the importance of building ML-based systems with a purpose by focusing on the problem first. We will see the current status of ML applications, the adoption properties, and engineering mechanisms to ensure our ML projects align with the problems they are designed for.
• Data-Orientation
  Date: May 31, 2025 Description: This lecture will start looking into the data dimension of the ML concept and will emphasise on the importance of data-orientation. We first define the concept of data, the associated challenges, and provide examples of data collection processes. We then define a data science methodology to iteratively build the datasets that will feed our machine learning models. This lecture explores the first step of this methodology, "data access".
• Data Quality
  Date: June 07, 2025 Description: This lecture explores the crucial aspects of data quality in machine learning. We will cover data cleaning, preprocessing, and transformation techniques essential for preparing datasets for various ML algorithms. The lecture emphasizes practical approaches to handle missing values, outliers, data normalization, feature engineering, and data validation to ensure high-quality input for ML models.
• Learning from Data
  Date: June 14, 2025 Description: This lecture explores the idea of learning from data. We explore the induction process that agents follow to go from a specific set of observations to general rules. This process enables the agents to make predictions about the future based on past experiences. We introduce the main concepts around this idea and show its application to regression and classification problems.
• The Perceptron
  Date: June 21, 2025 Description: We will continue exploring linear regression and classification models, building upon these foundations to introduce the perceptron as a fundamental building block for neural networks. We examine their mathematical formulation, learning process, and limitations, concluding with how these simple units form the basis for neural network architectures.
• Neural Networks
  Date: June 28, 2025 Description: We will continue our course exploring neural networks. We will build on previous concepts to define and formalise these supervised model that constitute the foundations of the latest advances in Machine Learning. We will also introduce and formalise Deep Learning models, their architectures and implementation details.
• Reinforcement Learning
  Date: July 05, 2025 Description: In this lecture, we will explore a different approach to learn from feedback. Reinforcement Learning is an alternative, and sometimes a complement, to supervised learning in which autonomous agents learn from interacting with the environment where they act. We will explore the theoretical and practical foundations of this approach.
• The Transformer Architecture
  Date: July 12, 2025 Description: This lecture introduces the transformer neural network architecture, which is the architecture of novel Large Language Models (LLMs). We will start formalising the architecture and its training. We will then introduce how to use and tailor LLMs into our ML projects and daily activities for different purposes.
• AI Systems
  Date: July 19, 2025 Description: In our last lecture, we will explore the deployment process of ML models. Once models are trained, we need to deploy them as part of larger systems to be used (i.e., AI Systems). We will show the challenges that motivate our focus on this stage and present different alternatives for efficient AI systems deployment.