Navigation

13M031ADO - Algorithms for dynamical optimization

Course specification
Course title Algorithms for dynamical optimization
Acronym 13M031ADO
Study programme Electrical Engineering and Computing
Module System Engineering and Radio Communications
Type of study master academic studies
Lecturer (for classes)
Lecturer/Associate (for practice)
    Lecturer/Associate (for OTC)
      ESPB 6.0 Status elective
      Condition -
      The goal Introduction to the basic dynamic optimization algorithms and their aplications in information theory and telecommunications, as well as in other similar scientific fields where these algorithms are used, like machine learning and bionformatics.
      The outcome Students will learn the basic concepts of statistical decision-making by using iterative algorithms for dynamic optimization. They will also learn how to implement the described graphical models and algorithms and use them to solve various problems related to information transmition and processing.
      Contents
      Contents of lectures Modelling and factor graph based decomposition of optimization problems in engineering. Applications of finite state machines and Bayesian networks in information theory. Iterative learning on graphs. Belief propagation algorithm and its applications in iterative decoding. Viterbi and Baum-Welch methods with applications in turbo decoding and channel equalization. Gradient-based optimizations.
      Contents of exercises Software-based demonstrations of iterative dynamic optimization algorithms. Examples of practically significant optimization problems in information theory and related engineering fields. Homeworks that follow lecture topics.
      Literature
      1. D. J.C. MecKay, Information Theory, Inference, and Learning Algorithms, Cambridge University Press, 2003 (Original title)
      2. T. Richardson, R. Urbanke, Modern Coding Theory, Cambridge University Press, 2009 (Original title)
      3. T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Springer, 2009 (Original title)
      4. D. Drajić, P. Ivaniš, Introduction to Information Theory and Coding, IV ed, Akademska misao, Beograd, 2018.
      5. P. Ivanis, D. Drajic, Information Theory and Coding - Solved Problems, Springer, New York, 2017 (Original title)
      Number of hours per week during the semester/trimester/year
      Lectures Exercises OTC Study and Research Other classes
      2 2
      Methods of teaching Teaching methods comprise lectures and precepts. Homeworks and student projects.
      Knowledge score (maximum points 100)
      Pre obligations Points Final exam Points
      Activites during lectures 0 Test paper 60
      Practical lessons 40 Oral examination 0
      Projects 0
      Colloquia 0
      Seminars 0