This is an archived version of the Machine Learning Lab website for documentation purposes. The host is in no way affiliated with the University of Freiburg, its Faculty of Engineering or any of its members.

Spezialvorlesung (Special Lecture) KI (E)

Optimierendes Lernen (Reinforcement Learning)
Dr. Joschka Bödecker, Jan Wülfing

  • Ankündigungen:
    • erste Vorlesung: 19.10.2015
    • Übungen im Wechsel mit Vorlesung nach Ankündigung
  • Vorlesungen/ Übungen
    • Montag, 14:00 - 16:00, Geb. 052 - SR 02-017
    • Mittwoch, 16:00 - 18:00, Geb. 052 - SR 02-017
    • Update: Vorlesung am 01.02.16 verschoben auf 03.02.16 12:30 - 14:00, Geb. 052 - SR 02-017
    • Update: Vorlesung am 08.02.16 verschoben auf 10.02.16 12:30 - 14:00, Geb. 052 - SR 02-017
  • Prüfung
    • Bachelor: Mündliche Prüfung
    • Master: Mündliche Prüfung
  • Kreditpunkte:
    • 6 ECTS
  • Sprache:
    • Englisch

Überblick:

The lecture deals with methods of Reinforcement Learning that constitute an important class of machine learning algorithms. Starting with the formalization of problems as Markov decision processes, a variety of Reinforcement Learning methods are introduced and discussed in-depth. The connection to practice-oriented problems is established throughout the lecture based on many examples.

Folien:

  • chapter 1 (introduction) PDF
  • chapter 2 (MDPs and dynamic programming) PDF
  • chapter 3+4 (value iteration + policy iteration) PDF
  • chapter 5 (modeling the world with MDPs) PDF
  • chapter 6 (TD-learning) PDF
  • chapter 7 (optimistic policy iteration) PDF
  • chapter 8 (trajectory based learning) PDF
  • chapter 9 (Q-learning) PDF
  • chapter 10 (function approximators) PDF
  • chapter 14 (RL with function approximators) PDF
  • chapter 15 (Policy gradient approaches) PDF
  • chapter 16 (wrap-up) PDF