SubjectsSubjects(version: 845)
Course, academic year 2019/2020
   Login via CAS
Reinforcement Learning and its Applications - NAIL117
Title in English: Posilované učení a jeho aplikace
Guaranteed by: Department of Theoretical Computer Science and Mathematical Logic (32-KTIML)
Faculty: Faculty of Mathematics and Physics
Actual: from 2018
Semester: winter
E-Credits: 3
Hours per week, examination: winter s.:1/1 C+Ex [hours/week]
Capacity: unlimited
Min. number of students: unlimited
State of the course: taught
Language: English
Teaching methods: full-time
Guarantor: Mgr. Karel Macek
Class: Informatika Mgr. - Teoretická informatika
Classification: Informatics > Informatics, Software Applications, Computer Graphics and Geometry, Database Systems, Didactics of Informatics, Discrete Mathematics, External Subjects, General Subjects, Computer and Formal Linguistics, Optimalization, Programming, Software Engineering, Theoretical Computer Science
Annotation -
Last update: RNDr. Jan Hric (27.04.2018)
The theory of Reinforcement Learning (RL) is motivated by results about the rational behavior of agents in dynamic environments and relates them to the context of machine learning, control theory, and statistical decision theory. The RL algorithms are applied in various fields, including control of physical systems or playing computer games.
Aim of the course -
Last update: RNDr. Jan Hric (27.04.2018)

To introduce the reinforcement learning principles and concepts, to explain fundamental algorithms, and practice them in Python.

Course completion requirements -
Last update: RNDr. Jan Hric (27.04.2018)

It is necessary to get course credit and pass the examination (in arbitrary order). Credit is given for solving homework, including potential additional homework at the end of the semester. The nature of this requirement excludes the possibility of repeated attempts to get credit, which means that if you do not obtain enough points for homework, there is no other way how to get credit.

The examination consists of a written and oral part. The written part precedes the oral part, a failure in the written part implies failing the whole exam, so the oral part is skipped in such cases.

Literature -
Last update: RNDr. Jan Hric (09.05.2018)

Richard S. Sutton, Andrew G. Barto. Reinforcement learning: An introduction. Vol. 1. No. 1. Cambridge: MIT Press, 1998.

Busoniu, L., Babuska, R., De Schutter, B., & Ernst, D. (2010). Reinforcement learning and dynamic programming using function approximators (Vol. 39). CRC press.

Syllabus -
Last update: RNDr. Jan Hric (30.04.2018)

Introduction to RL

  • basic elements, terminology
  • motivational examples

Tabular methods

  • multiarmed bandits
  • finite Markov decision processes
  • dynamic programming
  • Monte Carlo methods
  • TD methods

Approximation

  • categories of approximation methods
  • application of neural networks

The course is given (only) in English.

 
Charles University | Information system of Charles University | http://www.cuni.cz/UKEN-329.html