Advanced machine learning methods and their application in games

• Thesis title in English: Advanced machine learning methods and their application in games
• Key words: Strojové učení, Hry, Umělá inteligence
• English key words: Machine learning, Games, Artificial intelligence
• Academic year of topic announcement: 2013/2014
• Thesis type: dissertation
• Thesis language: angličtina
• Department: Department of Theoretical Computer Science and Mathematical Logic (32-KTIML)
• Supervisor: Mgr. Roman Neruda, CSc.
• Author: hidden - assigned and confirmed by the Study Dept.
• Date of registration: 26.09.2014
• Date of assignment: 26.09.2014
• Confirmed by Study dept. on: 26.01.2015

Guidelines

Games with large branching factors, such as Go or Arimaa, pose a challenging test-ground for the AI research. The techniques developed for solving such hard problems have often been found widely applicable outside the AI. The goal of this work will be to study and create machine learning algorithms with multiple possible goals: improving the performance of AI solvers for such games, improving the user experience when playing the AI bots, or helping the user himself to improve.
Part of the work will focus on researching the current trends and state of the art methods. Expected results include design and implementation of algorithms based on hybrid machine learning methods, including meta-learning approaches. The student will present the findings at international conferences and in peer-reviewed journals.

References

[1] Stuart J. Russell, Peter Norvig: Artificial Intelligence - A Modern Approach (3. internat. ed.), Pearson Education, 2010.

[2] Christopher M. Bishop, Nasser M. Nasrabadi: Pattern Recognition and Machine Learning. J. Electronic Imaging 16(4), 2007.

[3] Pavel Brazdil, Christophe G. Giraud-Carrier, Carlos Soares, Ricardo Vilalta: Metalearning - Applications to Data Mining. Cognitive Technologies, Springer, 2009.