Navigation

13M051MU - Machine Learning

Course specification
Course title Machine Learning
Acronym 13M051MU
Study programme Electrical Engineering and Computing
Module
Type of study master academic studies
Lecturer (for classes)
Lecturer/Associate (for practice)
Lecturer/Associate (for OTC)
    ESPB 6.0 Status elective
    Condition none
    The goal Introduce students to theoretical and practical aspects of supervised machine learning and reinforcement learning. illustration of various application areas, with guidelines on how to choose the adequate model, and how to optimize, evaluate and implement it.
    The outcome Students will be able to: choose an adequate machine learning algorithm suited to real-world problems, implement it, optimize its parameters and evaluate its performance. Special attention will be paid to techniques for formulating the problem and casting it in a configuration best suited to the application of the methods covered in this course.
    Contents
    URL to the subject page http://automatika.etf.bg.ac.rs/sr/13m051mu
    URL to lectures https://teams.microsoft.com/l/team/19%3a0z9Td-2NHnXtQpVG_yI1smbhT58-fU4RVik0apgX7AU1%40thread.tacv2/conversations?groupId=a2dfcaf8-af8d-487f-9d09-05abd6bf14cb&tenantId=1774ef2e-9c62-478a-8d3a-fd2a495547ba
    Contents of lectures Linear and logistic regression. Numerical optimization methods. Exponential family of distributions and generalized linear models. Generative algorithms. Support vector machines. Decision trees. Bagging, boosting, AdaBoost, Random Forests. Gaussian Processes. Model and feature selection. Learning theory: bias and variance, VC-dimension. Reinforcement learning.
    Contents of exercises Implementation of regressors and classifiers of simulated and real-world data using Python and Matlab/Octave. Implementation of reinforcement learning algorithms in simulated environments.
    Literature
    1. C. Bishop, "Pattern Recognition and Machine Learning", Springer, 2007 (Original title)
    2. Stuart Russel, Peter Norvig, "Artificial Intelligence: A Modern Approach", Pearson, 2010 (Original title)
    3. T. Hastie, R. Tibshirani, J. Friedman, "The Elements of Statistical Learning: Data Mining, Inference, and Prediction", Springer, 2008 (Original title)
    Number of hours per week during the semester/trimester/year
    Lectures Exercises OTC Study and Research Other classes
    3 1
    Methods of teaching Lectures, recitals, homework and project.
    Knowledge score (maximum points 100)
    Pre obligations Points Final exam Points
    Activites during lectures Test paper 60
    Practical lessons 20 Oral examination
    Projects 20
    Colloquia
    Seminars