Università Roma Tre

1. Machine learning. Types of learning. Loss functions. Empirical risk minimization. Generalization and overfitting.
2. Mathematical optimization problems. Convex functions. Gradient descent. Stochastic gradient descent.
3. Regression. Linear regression. Basis functions. Feature selection. Regularization.
4. Classification. Generative models. Nearest neighbor. Logistic regression. Support vector machines. Neural networks.
5. Combining models. Decision trees. Boosting.
6. Unsupervised learning. K-means clustering. Hierarchical clustering. Principal component analysis.
7. Application of the methods using the Python language. Examples using the NumPy, Pandas, SciKit-Learn, Keras, and TensorFlow libraries.

J. Watt, R. Borhani, A. Katsaggelos. Machine Learning Refined. Cambridge University Press, 2nd edition, 2020.