1. komponenta

Lecture type	Total
Lectures	45
Exercises	30

* Load is given in academic hour (1 academic hour = 45 minutes)

COURSE AIMS AND OBJECTIVES:
Students will become familiar with:
- the problem of supervised/unsupervised learning;
- basic machine learning algorithms and learn how to solve them using numerical optimization;
- the theoretical background of these algorithms using the framework of statistical learning theory.

COURSE DESCRIPTION AND SYLLABUS:
We will cover topics which might include:

1. Notions: bias-variance trade off, generalization error, model selection, variable selection, cross-validation, bootstrap, regularization, optimization for machine learning.
2. Supervised learning methods: linear models, penalized linear models, basic classifications methods such as logistic regression, LDA/QDA and naive Bayes, local methods, SVM, neural networks.
3. Unsupervised learning methods: principal component analysis, clustering methods.

1. The Elements of Statistical Learning: Data Mining, Inference and Prediction, T. Hastie, R. Tibshirani, and J. Friedman, Springer, 2009.
2. Deep learning, I. Goodfellow, Y. Bengio, A. Courville, MIT press, 2016.
3. Learning theory from first principles, F. Bach, MIT press, 2024.
4. All of Statistics: A Concise Course in Statistical Inference, L. Wasserman, Springer, 2004.
5. Mathematics for Machine Learning, M. P. Deisenroth, A. A. Faisal, C. S. Ong, Cambridge University Press, 2020.
6. Pattern Recognition and Machine Learning, C. Bishop, Springer, 2007.
7. Machine Learning: a Probabilistic Perspective, K. Murphy, MIT, 2012.

Mandatory course - Regular study - Computer Science and Mathematics