Instructor for half of the class (lectures and technical labs, 22.5 hours, spring 2026). Introduction to Machine and Deep Learning, with industrial applications.

Contents

• Supervised learning and kNNs
• Tree-based methods (Decision Trees)
• Data preprocessing
• Model evaluation (over-fitting, cross-validation, model-selection, performance metrics)
• Single-layer neural networks and the perceptron algorithm
• Parameter optimization with gradient descent, automatic differentiation with DL frameworks
• Neural Networks: Multinomial logistic regression / Softmax regression, Multilayer perceptrons and backpropagation, Regularization to avoid overfitting, Input normalization and weight initialization
• DL for computer vision and language modeling: Introduction to CNNs, RNNs, Autoencoders, Self-attention and transformer networks
• Applications to industrial data
• Preparation of a research project in the intersection of an Industrial Application and ML