Fortgeschrittene Methoden des maschinellen Lernens für Simulationsfußball
(Advanced Machine Learning Methods for Soccer Simulation)
Prof. Dr. M. Riedmiller and Dr. Thomas Gabel

• Meetings:
  • Tuesdays, 14.15-15.45
  • room SR00029 (pool), building 082
• Start:
  • first meeting: 20.10.2009
• Credit Points:
  • 6 ECTS
• Requirements:
  • good programming skills (C/C++)
  • basic knowledge of machine learning
  • having attended the lecture Machine Learning (summer term 2009) or attending the lecture Optimierendes Lernen (Reinforcement Learning) (this term) is advantageous
• Language:
  • English

RoboCup is an international research initiative intending to expedite artificial intelligence and intelligent robotics research by defining a set of standard problems where various technologies can be combined solving them. Annually, there are championship tournaments in several leagues, ranging from rescue tasks over real soccer-playing robots to simulated ones. Traditionally, the soccer simulation league has been the most prominent section of RoboCup for investigating and applying AI-based and machine learning approaches for soccer-playing agents. In this practical course the participants will aim at the development of machine learning approaches and at their deployment in the scope of a robotic soccer simulation team.

Follow this link to retrieve all materials related to this course (login data will be provided during the first meeting).