Seminar on machine learning with Simone Callegari

The seminar offers a structured and accessible overview of the most widely used machine learning algorithms. We will explore both supervised and unsupervised learning paradigms, highlighting key methods such as regression techniques, tree-based algorithms, support vector machines, k-nearest neighbors, naïve Bayes, k-means clustering, principal component analysis and artificial neural networks.

The session will also cover:

• The machine learning pipeline: from training and inference to validation and cross-validation
• Core concepts in classification, regression, clustering, and feature extraction
• Evaluation metrics such as accuracy, precision, recall, F1 score, and R²
• Common pitfalls like overfitting and underfitting, and how to mitigate them

The seminar is designed for an interdisciplinary audience, the presentation emphasizes conceptual clarity and practical relevance, equipping participants with a foundational understanding of how machine learning can support predictive analytics and decision-making.